diff --git "a/annotation_CSV/PMC4848761.csv" "b/annotation_CSV/PMC4848761.csv"
new file mode 100644--- /dev/null
+++ "b/annotation_CSV/PMC4848761.csv"
@@ -0,0 +1,884 @@
+anno_start	anno_end	anno_text	entity_type	sentence	section
+61	78	estrogen receptor	protein_type	Predictive features of ligand‐specific signaling through the estrogen receptor	TITLE
+5	24	estrogen receptor‐α	protein	Some estrogen receptor‐α (ERα)‐targeted breast cancer therapies such as tamoxifen have tissue‐selective or cell‐specific activities, while others have similar activities in different cell types.	ABSTRACT
+26	29	ERα	protein	Some estrogen receptor‐α (ERα)‐targeted breast cancer therapies such as tamoxifen have tissue‐selective or cell‐specific activities, while others have similar activities in different cell types.	ABSTRACT
+72	81	tamoxifen	chemical	Some estrogen receptor‐α (ERα)‐targeted breast cancer therapies such as tamoxifen have tissue‐selective or cell‐specific activities, while others have similar activities in different cell types.	ABSTRACT
+105	116	synthesized	experimental_method	To identify biophysical determinants of cell‐specific signaling and breast cancer cell proliferation, we synthesized 241 ERα ligands based on 19 chemical scaffolds, and compared ligand response using quantitative bioassays for canonical ERα activities and X‐ray crystallography.	ABSTRACT
+121	124	ERα	protein	To identify biophysical determinants of cell‐specific signaling and breast cancer cell proliferation, we synthesized 241 ERα ligands based on 19 chemical scaffolds, and compared ligand response using quantitative bioassays for canonical ERα activities and X‐ray crystallography.	ABSTRACT
+200	222	quantitative bioassays	experimental_method	To identify biophysical determinants of cell‐specific signaling and breast cancer cell proliferation, we synthesized 241 ERα ligands based on 19 chemical scaffolds, and compared ligand response using quantitative bioassays for canonical ERα activities and X‐ray crystallography.	ABSTRACT
+237	240	ERα	protein	To identify biophysical determinants of cell‐specific signaling and breast cancer cell proliferation, we synthesized 241 ERα ligands based on 19 chemical scaffolds, and compared ligand response using quantitative bioassays for canonical ERα activities and X‐ray crystallography.	ABSTRACT
+256	277	X‐ray crystallography	experimental_method	To identify biophysical determinants of cell‐specific signaling and breast cancer cell proliferation, we synthesized 241 ERα ligands based on 19 chemical scaffolds, and compared ligand response using quantitative bioassays for canonical ERα activities and X‐ray crystallography.	ABSTRACT
+56	80	coactivator‐binding site	site	Ligands that regulate the dynamics and stability of the coactivator‐binding site in the C‐terminal ligand‐binding domain, called activation function‐2 (AF‐2), showed similar activity profiles in different cell types.	ABSTRACT
+99	120	ligand‐binding domain	structure_element	Ligands that regulate the dynamics and stability of the coactivator‐binding site in the C‐terminal ligand‐binding domain, called activation function‐2 (AF‐2), showed similar activity profiles in different cell types.	ABSTRACT
+129	150	activation function‐2	structure_element	Ligands that regulate the dynamics and stability of the coactivator‐binding site in the C‐terminal ligand‐binding domain, called activation function‐2 (AF‐2), showed similar activity profiles in different cell types.	ABSTRACT
+152	156	AF‐2	structure_element	Ligands that regulate the dynamics and stability of the coactivator‐binding site in the C‐terminal ligand‐binding domain, called activation function‐2 (AF‐2), showed similar activity profiles in different cell types.	ABSTRACT
+134	143	NCOA1/2/3	protein	Such ligands induced breast cancer cell proliferation in a manner that was predicted by the canonical recruitment of the coactivators NCOA1/2/3 and induction of the GREB1 proliferative gene.	ABSTRACT
+165	170	GREB1	protein	Such ligands induced breast cancer cell proliferation in a manner that was predicted by the canonical recruitment of the coactivators NCOA1/2/3 and induction of the GREB1 proliferative gene.	ABSTRACT
+33	54	inter‐atomic distance	evidence	For some ligand series, a single inter‐atomic distance in the ligand‐binding domain predicted their proliferative effects.	ABSTRACT
+62	83	ligand‐binding domain	structure_element	For some ligand series, a single inter‐atomic distance in the ligand‐binding domain predicted their proliferative effects.	ABSTRACT
+28	52	coactivator‐binding site	site	In contrast, the N‐terminal coactivator‐binding site, activation function‐1 (AF‐1), determined cell‐specific signaling induced by ligands that used alternate mechanisms to control cell proliferation.	ABSTRACT
+54	75	activation function‐1	structure_element	In contrast, the N‐terminal coactivator‐binding site, activation function‐1 (AF‐1), determined cell‐specific signaling induced by ligands that used alternate mechanisms to control cell proliferation.	ABSTRACT
+77	81	AF‐1	structure_element	In contrast, the N‐terminal coactivator‐binding site, activation function‐1 (AF‐1), determined cell‐specific signaling induced by ligands that used alternate mechanisms to control cell proliferation.	ABSTRACT
+20	47	systems structural analyses	experimental_method	Thus, incorporating systems structural analyses with quantitative chemical biology reveals how ligands can achieve distinct allosteric signaling outcomes through ERα.	ABSTRACT
+53	82	quantitative chemical biology	experimental_method	Thus, incorporating systems structural analyses with quantitative chemical biology reveals how ligands can achieve distinct allosteric signaling outcomes through ERα.	ABSTRACT
+162	165	ERα	protein	Thus, incorporating systems structural analyses with quantitative chemical biology reveals how ligands can achieve distinct allosteric signaling outcomes through ERα.	ABSTRACT
+82	109	G protein‐coupled receptors	protein_type	Many drugs are small‐molecule ligands of allosteric signaling proteins, including G protein‐coupled receptors (GPCRs) and nuclear receptors such as ERα.	INTRO
+111	116	GPCRs	protein_type	Many drugs are small‐molecule ligands of allosteric signaling proteins, including G protein‐coupled receptors (GPCRs) and nuclear receptors such as ERα.	INTRO
+122	139	nuclear receptors	protein_type	Many drugs are small‐molecule ligands of allosteric signaling proteins, including G protein‐coupled receptors (GPCRs) and nuclear receptors such as ERα.	INTRO
+148	151	ERα	protein	Many drugs are small‐molecule ligands of allosteric signaling proteins, including G protein‐coupled receptors (GPCRs) and nuclear receptors such as ERα.	INTRO
+146	165	ligand‐binding site	site	Small‐molecule ligands control receptor activity by modulating recruitment of effector enzymes to distal regions of the receptor, relative to the ligand‐binding site.	INTRO
+23	51	estrogen receptor modulators	protein_type	For example, selective estrogen receptor modulators (SERMs) such as tamoxifen (Nolvadex®; AstraZeneca) or raloxifene (Evista®; Eli Lilly) (Fig 1A) block the ERα‐mediated proliferative effects of the native estrogen, 17β‐estradiol (E2), on breast cancer cells, but promote beneficial estrogenic effects on bone mineral density and adverse estrogenic effects such as uterine proliferation, fatty liver, or stroke (Frolik et al, 1996; Fisher et al, 1998; McDonnell et al, 2002; Jordan, 2003).	INTRO
+53	58	SERMs	protein_type	For example, selective estrogen receptor modulators (SERMs) such as tamoxifen (Nolvadex®; AstraZeneca) or raloxifene (Evista®; Eli Lilly) (Fig 1A) block the ERα‐mediated proliferative effects of the native estrogen, 17β‐estradiol (E2), on breast cancer cells, but promote beneficial estrogenic effects on bone mineral density and adverse estrogenic effects such as uterine proliferation, fatty liver, or stroke (Frolik et al, 1996; Fisher et al, 1998; McDonnell et al, 2002; Jordan, 2003).	INTRO
+68	77	tamoxifen	chemical	For example, selective estrogen receptor modulators (SERMs) such as tamoxifen (Nolvadex®; AstraZeneca) or raloxifene (Evista®; Eli Lilly) (Fig 1A) block the ERα‐mediated proliferative effects of the native estrogen, 17β‐estradiol (E2), on breast cancer cells, but promote beneficial estrogenic effects on bone mineral density and adverse estrogenic effects such as uterine proliferation, fatty liver, or stroke (Frolik et al, 1996; Fisher et al, 1998; McDonnell et al, 2002; Jordan, 2003).	INTRO
+79	88	Nolvadex®	chemical	For example, selective estrogen receptor modulators (SERMs) such as tamoxifen (Nolvadex®; AstraZeneca) or raloxifene (Evista®; Eli Lilly) (Fig 1A) block the ERα‐mediated proliferative effects of the native estrogen, 17β‐estradiol (E2), on breast cancer cells, but promote beneficial estrogenic effects on bone mineral density and adverse estrogenic effects such as uterine proliferation, fatty liver, or stroke (Frolik et al, 1996; Fisher et al, 1998; McDonnell et al, 2002; Jordan, 2003).	INTRO
+106	116	raloxifene	chemical	For example, selective estrogen receptor modulators (SERMs) such as tamoxifen (Nolvadex®; AstraZeneca) or raloxifene (Evista®; Eli Lilly) (Fig 1A) block the ERα‐mediated proliferative effects of the native estrogen, 17β‐estradiol (E2), on breast cancer cells, but promote beneficial estrogenic effects on bone mineral density and adverse estrogenic effects such as uterine proliferation, fatty liver, or stroke (Frolik et al, 1996; Fisher et al, 1998; McDonnell et al, 2002; Jordan, 2003).	INTRO
+118	125	Evista®	chemical	For example, selective estrogen receptor modulators (SERMs) such as tamoxifen (Nolvadex®; AstraZeneca) or raloxifene (Evista®; Eli Lilly) (Fig 1A) block the ERα‐mediated proliferative effects of the native estrogen, 17β‐estradiol (E2), on breast cancer cells, but promote beneficial estrogenic effects on bone mineral density and adverse estrogenic effects such as uterine proliferation, fatty liver, or stroke (Frolik et al, 1996; Fisher et al, 1998; McDonnell et al, 2002; Jordan, 2003).	INTRO
+157	160	ERα	protein	For example, selective estrogen receptor modulators (SERMs) such as tamoxifen (Nolvadex®; AstraZeneca) or raloxifene (Evista®; Eli Lilly) (Fig 1A) block the ERα‐mediated proliferative effects of the native estrogen, 17β‐estradiol (E2), on breast cancer cells, but promote beneficial estrogenic effects on bone mineral density and adverse estrogenic effects such as uterine proliferation, fatty liver, or stroke (Frolik et al, 1996; Fisher et al, 1998; McDonnell et al, 2002; Jordan, 2003).	INTRO
+206	214	estrogen	chemical	For example, selective estrogen receptor modulators (SERMs) such as tamoxifen (Nolvadex®; AstraZeneca) or raloxifene (Evista®; Eli Lilly) (Fig 1A) block the ERα‐mediated proliferative effects of the native estrogen, 17β‐estradiol (E2), on breast cancer cells, but promote beneficial estrogenic effects on bone mineral density and adverse estrogenic effects such as uterine proliferation, fatty liver, or stroke (Frolik et al, 1996; Fisher et al, 1998; McDonnell et al, 2002; Jordan, 2003).	INTRO
+216	229	17β‐estradiol	chemical	For example, selective estrogen receptor modulators (SERMs) such as tamoxifen (Nolvadex®; AstraZeneca) or raloxifene (Evista®; Eli Lilly) (Fig 1A) block the ERα‐mediated proliferative effects of the native estrogen, 17β‐estradiol (E2), on breast cancer cells, but promote beneficial estrogenic effects on bone mineral density and adverse estrogenic effects such as uterine proliferation, fatty liver, or stroke (Frolik et al, 1996; Fisher et al, 1998; McDonnell et al, 2002; Jordan, 2003).	INTRO
+231	233	E2	chemical	For example, selective estrogen receptor modulators (SERMs) such as tamoxifen (Nolvadex®; AstraZeneca) or raloxifene (Evista®; Eli Lilly) (Fig 1A) block the ERα‐mediated proliferative effects of the native estrogen, 17β‐estradiol (E2), on breast cancer cells, but promote beneficial estrogenic effects on bone mineral density and adverse estrogenic effects such as uterine proliferation, fatty liver, or stroke (Frolik et al, 1996; Fisher et al, 1998; McDonnell et al, 2002; Jordan, 2003).	INTRO
+22	25	ERα	protein	Allosteric control of ERα activity	FIG
+35	38	ERα	protein	Chemical structures of some common ERα ligands.	FIG
+0	2	E2	chemical	E2‐rings are numbered A‐D. The E‐ring is the common site of attachment for BSC found in many SERMS.	FIG
+93	98	SERMS	protein_type	E2‐rings are numbered A‐D. The E‐ring is the common site of attachment for BSC found in many SERMS.	FIG
+0	3	ERα	protein	ERα domain organization lettered, A‐F. DBD, DNA‐binding domain; LBD, ligand‐binding domain; AF, activation function	FIG
+39	42	DBD	structure_element	ERα domain organization lettered, A‐F. DBD, DNA‐binding domain; LBD, ligand‐binding domain; AF, activation function	FIG
+44	62	DNA‐binding domain	structure_element	ERα domain organization lettered, A‐F. DBD, DNA‐binding domain; LBD, ligand‐binding domain; AF, activation function	FIG
+64	67	LBD	structure_element	ERα domain organization lettered, A‐F. DBD, DNA‐binding domain; LBD, ligand‐binding domain; AF, activation function	FIG
+69	90	ligand‐binding domain	structure_element	ERα domain organization lettered, A‐F. DBD, DNA‐binding domain; LBD, ligand‐binding domain; AF, activation function	FIG
+92	94	AF	structure_element	ERα domain organization lettered, A‐F. DBD, DNA‐binding domain; LBD, ligand‐binding domain; AF, activation function	FIG
+96	115	activation function	structure_element	ERα domain organization lettered, A‐F. DBD, DNA‐binding domain; LBD, ligand‐binding domain; AF, activation function	FIG
+40	43	ERα	protein	Schematic illustration of the canonical ERα signaling pathway.	FIG
+27	30	ERα	protein	Linear causality model for ERα‐mediated cell proliferation.	FIG
+29	32	ERα	protein	Branched causality model for ERα‐mediated cell proliferation.	FIG
+0	3	ERα	protein	ERα contains structurally conserved globular domains of the nuclear receptor superfamily, including a DNA‐binding domain (DBD) that is connected by a flexible hinge region to the ligand‐binding domain (LBD), as well as unstructured AB and F domains at its amino and carboxyl termini, respectively (Fig 1B).	INTRO
+13	35	structurally conserved	protein_state	ERα contains structurally conserved globular domains of the nuclear receptor superfamily, including a DNA‐binding domain (DBD) that is connected by a flexible hinge region to the ligand‐binding domain (LBD), as well as unstructured AB and F domains at its amino and carboxyl termini, respectively (Fig 1B).	INTRO
+36	52	globular domains	structure_element	ERα contains structurally conserved globular domains of the nuclear receptor superfamily, including a DNA‐binding domain (DBD) that is connected by a flexible hinge region to the ligand‐binding domain (LBD), as well as unstructured AB and F domains at its amino and carboxyl termini, respectively (Fig 1B).	INTRO
+60	88	nuclear receptor superfamily	protein_type	ERα contains structurally conserved globular domains of the nuclear receptor superfamily, including a DNA‐binding domain (DBD) that is connected by a flexible hinge region to the ligand‐binding domain (LBD), as well as unstructured AB and F domains at its amino and carboxyl termini, respectively (Fig 1B).	INTRO
+102	120	DNA‐binding domain	structure_element	ERα contains structurally conserved globular domains of the nuclear receptor superfamily, including a DNA‐binding domain (DBD) that is connected by a flexible hinge region to the ligand‐binding domain (LBD), as well as unstructured AB and F domains at its amino and carboxyl termini, respectively (Fig 1B).	INTRO
+122	125	DBD	structure_element	ERα contains structurally conserved globular domains of the nuclear receptor superfamily, including a DNA‐binding domain (DBD) that is connected by a flexible hinge region to the ligand‐binding domain (LBD), as well as unstructured AB and F domains at its amino and carboxyl termini, respectively (Fig 1B).	INTRO
+150	158	flexible	protein_state	ERα contains structurally conserved globular domains of the nuclear receptor superfamily, including a DNA‐binding domain (DBD) that is connected by a flexible hinge region to the ligand‐binding domain (LBD), as well as unstructured AB and F domains at its amino and carboxyl termini, respectively (Fig 1B).	INTRO
+159	171	hinge region	structure_element	ERα contains structurally conserved globular domains of the nuclear receptor superfamily, including a DNA‐binding domain (DBD) that is connected by a flexible hinge region to the ligand‐binding domain (LBD), as well as unstructured AB and F domains at its amino and carboxyl termini, respectively (Fig 1B).	INTRO
+179	200	ligand‐binding domain	structure_element	ERα contains structurally conserved globular domains of the nuclear receptor superfamily, including a DNA‐binding domain (DBD) that is connected by a flexible hinge region to the ligand‐binding domain (LBD), as well as unstructured AB and F domains at its amino and carboxyl termini, respectively (Fig 1B).	INTRO
+202	205	LBD	structure_element	ERα contains structurally conserved globular domains of the nuclear receptor superfamily, including a DNA‐binding domain (DBD) that is connected by a flexible hinge region to the ligand‐binding domain (LBD), as well as unstructured AB and F domains at its amino and carboxyl termini, respectively (Fig 1B).	INTRO
+219	231	unstructured	protein_state	ERα contains structurally conserved globular domains of the nuclear receptor superfamily, including a DNA‐binding domain (DBD) that is connected by a flexible hinge region to the ligand‐binding domain (LBD), as well as unstructured AB and F domains at its amino and carboxyl termini, respectively (Fig 1B).	INTRO
+232	234	AB	structure_element	ERα contains structurally conserved globular domains of the nuclear receptor superfamily, including a DNA‐binding domain (DBD) that is connected by a flexible hinge region to the ligand‐binding domain (LBD), as well as unstructured AB and F domains at its amino and carboxyl termini, respectively (Fig 1B).	INTRO
+239	240	F	structure_element	ERα contains structurally conserved globular domains of the nuclear receptor superfamily, including a DNA‐binding domain (DBD) that is connected by a flexible hinge region to the ligand‐binding domain (LBD), as well as unstructured AB and F domains at its amino and carboxyl termini, respectively (Fig 1B).	INTRO
+4	7	LBD	structure_element	The LBD contains a ligand‐dependent coactivator‐binding site called activation function‐2 (AF‐2).	INTRO
+36	60	coactivator‐binding site	site	The LBD contains a ligand‐dependent coactivator‐binding site called activation function‐2 (AF‐2).	INTRO
+68	89	activation function‐2	structure_element	The LBD contains a ligand‐dependent coactivator‐binding site called activation function‐2 (AF‐2).	INTRO
+91	95	AF‐2	structure_element	The LBD contains a ligand‐dependent coactivator‐binding site called activation function‐2 (AF‐2).	INTRO
+33	38	SERMs	protein_type	However, the agonist activity of SERMs derives from activation function‐1 (AF‐1)—a coactivator recruitment site located in the AB domain (Berry et al, 1990; Shang & Brown, 2002; Abot et al, 2013).	INTRO
+52	73	activation function‐1	structure_element	However, the agonist activity of SERMs derives from activation function‐1 (AF‐1)—a coactivator recruitment site located in the AB domain (Berry et al, 1990; Shang & Brown, 2002; Abot et al, 2013).	INTRO
+75	79	AF‐1	structure_element	However, the agonist activity of SERMs derives from activation function‐1 (AF‐1)—a coactivator recruitment site located in the AB domain (Berry et al, 1990; Shang & Brown, 2002; Abot et al, 2013).	INTRO
+83	111	coactivator recruitment site	site	However, the agonist activity of SERMs derives from activation function‐1 (AF‐1)—a coactivator recruitment site located in the AB domain (Berry et al, 1990; Shang & Brown, 2002; Abot et al, 2013).	INTRO
+127	129	AB	structure_element	However, the agonist activity of SERMs derives from activation function‐1 (AF‐1)—a coactivator recruitment site located in the AB domain (Berry et al, 1990; Shang & Brown, 2002; Abot et al, 2013).	INTRO
+0	4	AF‐1	structure_element	AF‐1 and AF‐2 bind distinct but overlapping sets of coregulators (Webb et al, 1998; Endoh et al, 1999; Delage‐Mourroux et al, 2000; Yi et al, 2015).	INTRO
+9	13	AF‐2	structure_element	AF‐1 and AF‐2 bind distinct but overlapping sets of coregulators (Webb et al, 1998; Endoh et al, 1999; Delage‐Mourroux et al, 2000; Yi et al, 2015).	INTRO
+0	4	AF‐2	structure_element	AF‐2 binds the signature LxxLL motif peptides of coactivators such as NCOA1/2/3 (also known as SRC‐1/2/3).	INTRO
+25	36	LxxLL motif	structure_element	AF‐2 binds the signature LxxLL motif peptides of coactivators such as NCOA1/2/3 (also known as SRC‐1/2/3).	INTRO
+70	79	NCOA1/2/3	protein	AF‐2 binds the signature LxxLL motif peptides of coactivators such as NCOA1/2/3 (also known as SRC‐1/2/3).	INTRO
+95	104	SRC‐1/2/3	protein	AF‐2 binds the signature LxxLL motif peptides of coactivators such as NCOA1/2/3 (also known as SRC‐1/2/3).	INTRO
+0	4	AF‐1	structure_element	AF‐1 binds a separate surface on these coactivators (Webb et al, 1998; Yi et al, 2015).	INTRO
+33	36	ERα	protein	Yet, it is unknown how different ERα ligands control AF‐1 through the LBD, and whether this inter‐domain communication is required for cell‐specific signaling or anti‐proliferative responses.	INTRO
+53	57	AF‐1	structure_element	Yet, it is unknown how different ERα ligands control AF‐1 through the LBD, and whether this inter‐domain communication is required for cell‐specific signaling or anti‐proliferative responses.	INTRO
+70	73	LBD	structure_element	Yet, it is unknown how different ERα ligands control AF‐1 through the LBD, and whether this inter‐domain communication is required for cell‐specific signaling or anti‐proliferative responses.	INTRO
+30	33	ERα	protein	In the canonical model of the ERα signaling pathway (Fig 1C), E2‐bound ERα forms a homodimer that binds DNA at estrogen‐response elements (EREs), recruits NCOA1/2/3 (Metivier et al, 2003; Johnson & O'Malley, 2012), and activates the GREB1 gene, which is required for proliferation of ERα‐positive breast cancer cells (Ghosh et al, 2000; Rae et al, 2005; Deschenes et al, 2007; Liu et al, 2012; Srinivasan et al, 2013).	INTRO
+62	70	E2‐bound	protein_state	In the canonical model of the ERα signaling pathway (Fig 1C), E2‐bound ERα forms a homodimer that binds DNA at estrogen‐response elements (EREs), recruits NCOA1/2/3 (Metivier et al, 2003; Johnson & O'Malley, 2012), and activates the GREB1 gene, which is required for proliferation of ERα‐positive breast cancer cells (Ghosh et al, 2000; Rae et al, 2005; Deschenes et al, 2007; Liu et al, 2012; Srinivasan et al, 2013).	INTRO
+71	74	ERα	protein	In the canonical model of the ERα signaling pathway (Fig 1C), E2‐bound ERα forms a homodimer that binds DNA at estrogen‐response elements (EREs), recruits NCOA1/2/3 (Metivier et al, 2003; Johnson & O'Malley, 2012), and activates the GREB1 gene, which is required for proliferation of ERα‐positive breast cancer cells (Ghosh et al, 2000; Rae et al, 2005; Deschenes et al, 2007; Liu et al, 2012; Srinivasan et al, 2013).	INTRO
+83	92	homodimer	oligomeric_state	In the canonical model of the ERα signaling pathway (Fig 1C), E2‐bound ERα forms a homodimer that binds DNA at estrogen‐response elements (EREs), recruits NCOA1/2/3 (Metivier et al, 2003; Johnson & O'Malley, 2012), and activates the GREB1 gene, which is required for proliferation of ERα‐positive breast cancer cells (Ghosh et al, 2000; Rae et al, 2005; Deschenes et al, 2007; Liu et al, 2012; Srinivasan et al, 2013).	INTRO
+111	137	estrogen‐response elements	site	In the canonical model of the ERα signaling pathway (Fig 1C), E2‐bound ERα forms a homodimer that binds DNA at estrogen‐response elements (EREs), recruits NCOA1/2/3 (Metivier et al, 2003; Johnson & O'Malley, 2012), and activates the GREB1 gene, which is required for proliferation of ERα‐positive breast cancer cells (Ghosh et al, 2000; Rae et al, 2005; Deschenes et al, 2007; Liu et al, 2012; Srinivasan et al, 2013).	INTRO
+139	143	EREs	site	In the canonical model of the ERα signaling pathway (Fig 1C), E2‐bound ERα forms a homodimer that binds DNA at estrogen‐response elements (EREs), recruits NCOA1/2/3 (Metivier et al, 2003; Johnson & O'Malley, 2012), and activates the GREB1 gene, which is required for proliferation of ERα‐positive breast cancer cells (Ghosh et al, 2000; Rae et al, 2005; Deschenes et al, 2007; Liu et al, 2012; Srinivasan et al, 2013).	INTRO
+155	164	NCOA1/2/3	protein	In the canonical model of the ERα signaling pathway (Fig 1C), E2‐bound ERα forms a homodimer that binds DNA at estrogen‐response elements (EREs), recruits NCOA1/2/3 (Metivier et al, 2003; Johnson & O'Malley, 2012), and activates the GREB1 gene, which is required for proliferation of ERα‐positive breast cancer cells (Ghosh et al, 2000; Rae et al, 2005; Deschenes et al, 2007; Liu et al, 2012; Srinivasan et al, 2013).	INTRO
+233	238	GREB1	protein	In the canonical model of the ERα signaling pathway (Fig 1C), E2‐bound ERα forms a homodimer that binds DNA at estrogen‐response elements (EREs), recruits NCOA1/2/3 (Metivier et al, 2003; Johnson & O'Malley, 2012), and activates the GREB1 gene, which is required for proliferation of ERα‐positive breast cancer cells (Ghosh et al, 2000; Rae et al, 2005; Deschenes et al, 2007; Liu et al, 2012; Srinivasan et al, 2013).	INTRO
+284	287	ERα	protein	In the canonical model of the ERα signaling pathway (Fig 1C), E2‐bound ERα forms a homodimer that binds DNA at estrogen‐response elements (EREs), recruits NCOA1/2/3 (Metivier et al, 2003; Johnson & O'Malley, 2012), and activates the GREB1 gene, which is required for proliferation of ERα‐positive breast cancer cells (Ghosh et al, 2000; Rae et al, 2005; Deschenes et al, 2007; Liu et al, 2012; Srinivasan et al, 2013).	INTRO
+9	12	ERα	protein	However, ERα‐mediated proliferative responses vary in a ligand‐dependent manner (Srinivasan et al, 2013); thus, it is not known whether this canonical model is widely applicable across diverse ERα ligands.	INTRO
+193	196	ERα	protein	However, ERα‐mediated proliferative responses vary in a ligand‐dependent manner (Srinivasan et al, 2013); thus, it is not known whether this canonical model is widely applicable across diverse ERα ligands.	INTRO
+131	148	crystal structure	evidence	Our long‐term goal is to be able to predict proliferative or anti‐proliferative activity of a ligand in different tissues from its crystal structure by identifying different structural perturbations that lead to specific signaling outcomes.	INTRO
+119	128	NCOA1/2/3	protein	The simplest response model for ligand‐specific proliferative effects is a linear causality model, where the degree of NCOA1/2/3 recruitment determines GREB1 expression, which in turn drives ligand‐specific cell proliferation (Fig 1D).	INTRO
+152	157	GREB1	protein	The simplest response model for ligand‐specific proliferative effects is a linear causality model, where the degree of NCOA1/2/3 recruitment determines GREB1 expression, which in turn drives ligand‐specific cell proliferation (Fig 1D).	INTRO
+126	129	LBD	structure_element	In this signaling model, multiple coregulator binding events and target genes (Won Jeong et al, 2012; Nwachukwu et al, 2014), LBD conformation, nucleocytoplasmic shuttling, the occupancy and dynamics of DNA binding, and other biophysical features could contribute independently to cell proliferation (Lickwar et al, 2012).	INTRO
+65	68	ERα	protein	To test these signaling models, we profiled a diverse library of ERα ligands using systems biology approaches to X‐ray crystallography and chemical biology (Srinivasan et al, 2013), including a series of quantitative bioassays for ERα function that were statistically robust and reproducible, based on the Z’‐statistic (Fig EV1A and B; see Materials and Methods).	INTRO
+113	134	X‐ray crystallography	experimental_method	To test these signaling models, we profiled a diverse library of ERα ligands using systems biology approaches to X‐ray crystallography and chemical biology (Srinivasan et al, 2013), including a series of quantitative bioassays for ERα function that were statistically robust and reproducible, based on the Z’‐statistic (Fig EV1A and B; see Materials and Methods).	INTRO
+139	155	chemical biology	experimental_method	To test these signaling models, we profiled a diverse library of ERα ligands using systems biology approaches to X‐ray crystallography and chemical biology (Srinivasan et al, 2013), including a series of quantitative bioassays for ERα function that were statistically robust and reproducible, based on the Z’‐statistic (Fig EV1A and B; see Materials and Methods).	INTRO
+231	234	ERα	protein	To test these signaling models, we profiled a diverse library of ERα ligands using systems biology approaches to X‐ray crystallography and chemical biology (Srinivasan et al, 2013), including a series of quantitative bioassays for ERα function that were statistically robust and reproducible, based on the Z’‐statistic (Fig EV1A and B; see Materials and Methods).	INTRO
+306	318	Z’‐statistic	evidence	To test these signaling models, we profiled a diverse library of ERα ligands using systems biology approaches to X‐ray crystallography and chemical biology (Srinivasan et al, 2013), including a series of quantitative bioassays for ERα function that were statistically robust and reproducible, based on the Z’‐statistic (Fig EV1A and B; see Materials and Methods).	INTRO
+8	18	determined	experimental_method	We also determined the structures of 76 distinct ERα LBD complexes bound to different ligand types, which allowed us to understand how diverse ligand scaffolds distort the active conformation of the ERα LBD.	INTRO
+23	33	structures	evidence	We also determined the structures of 76 distinct ERα LBD complexes bound to different ligand types, which allowed us to understand how diverse ligand scaffolds distort the active conformation of the ERα LBD.	INTRO
+49	52	ERα	protein	We also determined the structures of 76 distinct ERα LBD complexes bound to different ligand types, which allowed us to understand how diverse ligand scaffolds distort the active conformation of the ERα LBD.	INTRO
+53	56	LBD	structure_element	We also determined the structures of 76 distinct ERα LBD complexes bound to different ligand types, which allowed us to understand how diverse ligand scaffolds distort the active conformation of the ERα LBD.	INTRO
+67	75	bound to	protein_state	We also determined the structures of 76 distinct ERα LBD complexes bound to different ligand types, which allowed us to understand how diverse ligand scaffolds distort the active conformation of the ERα LBD.	INTRO
+172	178	active	protein_state	We also determined the structures of 76 distinct ERα LBD complexes bound to different ligand types, which allowed us to understand how diverse ligand scaffolds distort the active conformation of the ERα LBD.	INTRO
+199	202	ERα	protein	We also determined the structures of 76 distinct ERα LBD complexes bound to different ligand types, which allowed us to understand how diverse ligand scaffolds distort the active conformation of the ERα LBD.	INTRO
+203	206	LBD	structure_element	We also determined the structures of 76 distinct ERα LBD complexes bound to different ligand types, which allowed us to understand how diverse ligand scaffolds distort the active conformation of the ERα LBD.	INTRO
+297	306	estrogens	chemical	Our findings here indicate that specific structural perturbations can be tied to ligand‐selective domain usage and signaling patterns, thus providing a framework for structure‐based design of improved breast cancer therapeutics, and understanding the different phenotypic effects of environmental estrogens.	INTRO
+28	31	ERα	protein	High‐throughput screens for ERα ligand profiling	FIG
+11	34	ligand screening assays	experimental_method	Summary of ligand screening assays used to measure ER‐mediated activities.	FIG
+0	3	ERE	structure_element	ERE, estrogen‐response element; Luc, luciferase reporter gene; M2H, mammalian 2‐hybrid; UAS, upstream‐activating sequence.	FIG
+5	30	estrogen‐response element	structure_element	ERE, estrogen‐response element; Luc, luciferase reporter gene; M2H, mammalian 2‐hybrid; UAS, upstream‐activating sequence.	FIG
+32	35	Luc	experimental_method	ERE, estrogen‐response element; Luc, luciferase reporter gene; M2H, mammalian 2‐hybrid; UAS, upstream‐activating sequence.	FIG
+37	61	luciferase reporter gene	experimental_method	ERE, estrogen‐response element; Luc, luciferase reporter gene; M2H, mammalian 2‐hybrid; UAS, upstream‐activating sequence.	FIG
+63	66	M2H	experimental_method	ERE, estrogen‐response element; Luc, luciferase reporter gene; M2H, mammalian 2‐hybrid; UAS, upstream‐activating sequence.	FIG
+68	86	mammalian 2‐hybrid	experimental_method	ERE, estrogen‐response element; Luc, luciferase reporter gene; M2H, mammalian 2‐hybrid; UAS, upstream‐activating sequence.	FIG
+88	91	UAS	structure_element	ERE, estrogen‐response element; Luc, luciferase reporter gene; M2H, mammalian 2‐hybrid; UAS, upstream‐activating sequence.	FIG
+93	121	upstream‐activating sequence	structure_element	ERE, estrogen‐response element; Luc, luciferase reporter gene; M2H, mammalian 2‐hybrid; UAS, upstream‐activating sequence.	FIG
+12	16	AF‐1	structure_element	Strength of AF‐1 signaling does not determine cell‐specific signaling	RESULTS
+11	14	ERα	protein	To compare ERα signaling induced by diverse ligand types, we synthesized and assayed a library of 241 ERα ligands containing 19 distinct molecular scaffolds.	RESULTS
+61	84	synthesized and assayed	experimental_method	To compare ERα signaling induced by diverse ligand types, we synthesized and assayed a library of 241 ERα ligands containing 19 distinct molecular scaffolds.	RESULTS
+102	105	ERα	protein	To compare ERα signaling induced by diverse ligand types, we synthesized and assayed a library of 241 ERα ligands containing 19 distinct molecular scaffolds.	RESULTS
+50	54	lack	protein_state	These include 15 indirect modulator series, which lack a SERM‐like side chain and modulate coactivator binding indirectly from the ligand‐binding pocket (Fig 2A–E; Dataset EV1) (Zheng et al, 2012) (Zhu et al, 2012) (Muthyala et al, 2003; Seo et al, 2006) (Srinivasan et al, 2013) (Wang et al, 2012) (Liao et al, 2014) (Min et al, 2013).	RESULTS
+57	66	SERM‐like	protein_type	These include 15 indirect modulator series, which lack a SERM‐like side chain and modulate coactivator binding indirectly from the ligand‐binding pocket (Fig 2A–E; Dataset EV1) (Zheng et al, 2012) (Zhu et al, 2012) (Muthyala et al, 2003; Seo et al, 2006) (Srinivasan et al, 2013) (Wang et al, 2012) (Liao et al, 2014) (Min et al, 2013).	RESULTS
+131	152	ligand‐binding pocket	site	These include 15 indirect modulator series, which lack a SERM‐like side chain and modulate coactivator binding indirectly from the ligand‐binding pocket (Fig 2A–E; Dataset EV1) (Zheng et al, 2012) (Zhu et al, 2012) (Muthyala et al, 2003; Seo et al, 2006) (Srinivasan et al, 2013) (Wang et al, 2012) (Liao et al, 2014) (Min et al, 2013).	RESULTS
+95	98	h12	structure_element	We also generated four direct modulator series with side chains designed to directly dislocate h12 and thereby completely occlude the AF‐2 surface (Fig 2C and E; Dataset EV1) (Kieser et al, 2010).	RESULTS
+134	146	AF‐2 surface	site	We also generated four direct modulator series with side chains designed to directly dislocate h12 and thereby completely occlude the AF‐2 surface (Fig 2C and E; Dataset EV1) (Kieser et al, 2010).	RESULTS
+0	16	Ligand profiling	experimental_method	Ligand profiling using our quantitative bioassays revealed a wide range of ligand‐induced GREB1 expression, reporter gene activities, ERα‐coactivator interactions, and proliferative effects on MCF‐7 breast cancer cells (Figs EV1 and EV2A–J).	RESULTS
+27	49	quantitative bioassays	experimental_method	Ligand profiling using our quantitative bioassays revealed a wide range of ligand‐induced GREB1 expression, reporter gene activities, ERα‐coactivator interactions, and proliferative effects on MCF‐7 breast cancer cells (Figs EV1 and EV2A–J).	RESULTS
+90	95	GREB1	protein	Ligand profiling using our quantitative bioassays revealed a wide range of ligand‐induced GREB1 expression, reporter gene activities, ERα‐coactivator interactions, and proliferative effects on MCF‐7 breast cancer cells (Figs EV1 and EV2A–J).	RESULTS
+134	137	ERα	protein	Ligand profiling using our quantitative bioassays revealed a wide range of ligand‐induced GREB1 expression, reporter gene activities, ERα‐coactivator interactions, and proliferative effects on MCF‐7 breast cancer cells (Figs EV1 and EV2A–J).	RESULTS
+60	63	ERα	protein	This wide variance enabled us to probe specific features of ERα signaling using ligand class analyses, and identify signaling patterns shared by specific ligand series or scaffolds.	RESULTS
+80	101	ligand class analyses	experimental_method	This wide variance enabled us to probe specific features of ERα signaling using ligand class analyses, and identify signaling patterns shared by specific ligand series or scaffolds.	RESULTS
+28	31	ERα	protein	Classes of compounds in the ERα ligand library	FIG
+0	9	Structure	evidence	Structure of the E2‐bound ERα LBD in complex with an NCOA2 peptide of (PDB 1GWR).	FIG
+17	25	E2‐bound	protein_state	Structure of the E2‐bound ERα LBD in complex with an NCOA2 peptide of (PDB 1GWR).	FIG
+26	29	ERα	protein	Structure of the E2‐bound ERα LBD in complex with an NCOA2 peptide of (PDB 1GWR).	FIG
+30	33	LBD	structure_element	Structure of the E2‐bound ERα LBD in complex with an NCOA2 peptide of (PDB 1GWR).	FIG
+34	49	in complex with	protein_state	Structure of the E2‐bound ERα LBD in complex with an NCOA2 peptide of (PDB 1GWR).	FIG
+53	58	NCOA2	protein	Structure of the E2‐bound ERα LBD in complex with an NCOA2 peptide of (PDB 1GWR).	FIG
+26	29	ERα	protein	Structural details of the ERα LBD bound to the indicated ligands.	FIG
+30	33	LBD	structure_element	Structural details of the ERα LBD bound to the indicated ligands.	FIG
+34	42	bound to	protein_state	Structural details of the ERα LBD bound to the indicated ligands.	FIG
+7	9	E2	chemical	Unlike E2 (PDB 1GWR), TAM is a direct modulator with a BSC that dislocates h12 to block the NCOA2‐binding site (PDB 3ERT).	FIG
+22	25	TAM	chemical	Unlike E2 (PDB 1GWR), TAM is a direct modulator with a BSC that dislocates h12 to block the NCOA2‐binding site (PDB 3ERT).	FIG
+75	78	h12	structure_element	Unlike E2 (PDB 1GWR), TAM is a direct modulator with a BSC that dislocates h12 to block the NCOA2‐binding site (PDB 3ERT).	FIG
+92	110	NCOA2‐binding site	site	Unlike E2 (PDB 1GWR), TAM is a direct modulator with a BSC that dislocates h12 to block the NCOA2‐binding site (PDB 3ERT).	FIG
+0	4	OBHS	chemical	OBHS is an indirect modulator that dislocates the h11 C‐terminus to destabilize the h11–h12 interface (PDB 4ZN9).	FIG
+50	53	h11	structure_element	OBHS is an indirect modulator that dislocates the h11 C‐terminus to destabilize the h11–h12 interface (PDB 4ZN9).	FIG
+84	101	h11–h12 interface	site	OBHS is an indirect modulator that dislocates the h11 C‐terminus to destabilize the h11–h12 interface (PDB 4ZN9).	FIG
+4	7	ERα	protein	The ERα ligand library contains 241 ligands representing 15 indirect modulator scaffolds, plus 4 direct modulator scaffolds.	FIG
+0	3	ERα	protein	ERα ligands induced a range of agonist activity profiles	FIG
+52	57	GREB1	protein	To this end, we compared the average ligand‐induced GREB1 mRNA levels in MCF‐7 cells and 3×ERE‐Luc reporter gene activity in Ishikawa endometrial cancer cells (E‐Luc) or in HepG2 cells transfected with wild‐type ERα (L‐Luc ERα‐WT) (Figs 3A and EV2A–C).	RESULTS
+89	98	3×ERE‐Luc	experimental_method	To this end, we compared the average ligand‐induced GREB1 mRNA levels in MCF‐7 cells and 3×ERE‐Luc reporter gene activity in Ishikawa endometrial cancer cells (E‐Luc) or in HepG2 cells transfected with wild‐type ERα (L‐Luc ERα‐WT) (Figs 3A and EV2A–C).	RESULTS
+160	165	E‐Luc	experimental_method	To this end, we compared the average ligand‐induced GREB1 mRNA levels in MCF‐7 cells and 3×ERE‐Luc reporter gene activity in Ishikawa endometrial cancer cells (E‐Luc) or in HepG2 cells transfected with wild‐type ERα (L‐Luc ERα‐WT) (Figs 3A and EV2A–C).	RESULTS
+202	211	wild‐type	protein_state	To this end, we compared the average ligand‐induced GREB1 mRNA levels in MCF‐7 cells and 3×ERE‐Luc reporter gene activity in Ishikawa endometrial cancer cells (E‐Luc) or in HepG2 cells transfected with wild‐type ERα (L‐Luc ERα‐WT) (Figs 3A and EV2A–C).	RESULTS
+212	215	ERα	protein	To this end, we compared the average ligand‐induced GREB1 mRNA levels in MCF‐7 cells and 3×ERE‐Luc reporter gene activity in Ishikawa endometrial cancer cells (E‐Luc) or in HepG2 cells transfected with wild‐type ERα (L‐Luc ERα‐WT) (Figs 3A and EV2A–C).	RESULTS
+217	222	L‐Luc	experimental_method	To this end, we compared the average ligand‐induced GREB1 mRNA levels in MCF‐7 cells and 3×ERE‐Luc reporter gene activity in Ishikawa endometrial cancer cells (E‐Luc) or in HepG2 cells transfected with wild‐type ERα (L‐Luc ERα‐WT) (Figs 3A and EV2A–C).	RESULTS
+223	226	ERα	protein	To this end, we compared the average ligand‐induced GREB1 mRNA levels in MCF‐7 cells and 3×ERE‐Luc reporter gene activity in Ishikawa endometrial cancer cells (E‐Luc) or in HepG2 cells transfected with wild‐type ERα (L‐Luc ERα‐WT) (Figs 3A and EV2A–C).	RESULTS
+227	229	WT	protein_state	To this end, we compared the average ligand‐induced GREB1 mRNA levels in MCF‐7 cells and 3×ERE‐Luc reporter gene activity in Ishikawa endometrial cancer cells (E‐Luc) or in HepG2 cells transfected with wild‐type ERα (L‐Luc ERα‐WT) (Figs 3A and EV2A–C).	RESULTS
+95	103	OBHS‐ASC	chemical	Direct modulators showed significant differences in average activity between cell types except OBHS‐ASC analogs, which had similar low agonist activities in the three cell types.	RESULTS
+50	59	tamoxifen	chemical	While it was known that direct modulators such as tamoxifen drive cell‐specific signaling, these experiments reveal that indirect modulators also drive cell‐specific signaling, since eight of fourteen classes showed significant differences in average activity (Figs 3A and EV2A–C).	RESULTS
+36	39	ERα	protein	Ligand‐specific signaling underlies ERα‐mediated cell proliferation	FIG
+20	23	ERα	protein	(A) Ligand‐specific ERα activities in HepG2, Ishikawa and MCF‐7 cells.	FIG
+19	24	L‐Luc	experimental_method	The ligand‐induced L‐Luc ERα‐WT and E‐Luc activities and GREB1 mRNA levels are shown by scaffold (mean + SD).	FIG
+25	28	ERα	protein	The ligand‐induced L‐Luc ERα‐WT and E‐Luc activities and GREB1 mRNA levels are shown by scaffold (mean + SD).	FIG
+29	31	WT	protein_state	The ligand‐induced L‐Luc ERα‐WT and E‐Luc activities and GREB1 mRNA levels are shown by scaffold (mean + SD).	FIG
+36	41	E‐Luc	experimental_method	The ligand‐induced L‐Luc ERα‐WT and E‐Luc activities and GREB1 mRNA levels are shown by scaffold (mean + SD).	FIG
+57	62	GREB1	protein	The ligand‐induced L‐Luc ERα‐WT and E‐Luc activities and GREB1 mRNA levels are shown by scaffold (mean + SD).	FIG
+11	25	class analysis	experimental_method	(B) Ligand class analysis of the L‐Luc ERα‐WT and ERα‐ΔAB activities in HepG2 cells.	FIG
+33	38	L‐Luc	experimental_method	(B) Ligand class analysis of the L‐Luc ERα‐WT and ERα‐ΔAB activities in HepG2 cells.	FIG
+39	42	ERα	protein	(B) Ligand class analysis of the L‐Luc ERα‐WT and ERα‐ΔAB activities in HepG2 cells.	FIG
+43	45	WT	protein_state	(B) Ligand class analysis of the L‐Luc ERα‐WT and ERα‐ΔAB activities in HepG2 cells.	FIG
+50	57	ERα‐ΔAB	mutant	(B) Ligand class analysis of the L‐Luc ERα‐WT and ERα‐ΔAB activities in HepG2 cells.	FIG
+27	29	AB	structure_element	Significant sensitivity to AB domain deletion was determined by Student's t‐test (n = number of ligands per scaffold in Fig 2).	FIG
+64	80	Student's t‐test	experimental_method	Significant sensitivity to AB domain deletion was determined by Student's t‐test (n = number of ligands per scaffold in Fig 2).	FIG
+0	35	Correlation and regression analyses	experimental_method	Correlation and regression analyses in a large test set.	FIG
+91	97	F‐test	experimental_method	In cluster 1, the first three comparisons (rows) showed significant positive correlations (F‐test for nonzero slope, P ≤ 0.05).	FIG
+117	118	P	evidence	In cluster 1, the first three comparisons (rows) showed significant positive correlations (F‐test for nonzero slope, P ≤ 0.05).	FIG
+182	190	deletion	experimental_method	In cluster 2, only one of these comparisons revealed a significant positive correlation, while none was significant in cluster 3. +, statistically significant correlations gained by deletion of the AB or F domains.	FIG
+198	200	AB	structure_element	In cluster 2, only one of these comparisons revealed a significant positive correlation, while none was significant in cluster 3. +, statistically significant correlations gained by deletion of the AB or F domains.	FIG
+204	205	F	structure_element	In cluster 2, only one of these comparisons revealed a significant positive correlation, while none was significant in cluster 3. +, statistically significant correlations gained by deletion of the AB or F domains.	FIG
+50	52	AB	structure_element	−, significant correlations lost upon deletion of AB or F domains.	FIG
+56	57	F	structure_element	−, significant correlations lost upon deletion of AB or F domains.	FIG
+0	9	Tamoxifen	chemical	Tamoxifen depends on AF‐1 for its cell‐specific activity (Sakamoto et al, 2002); therefore, we asked whether cell‐specific signaling observed here is due to a similar dependence on AF‐1 for activity (Fig EV1).	RESULTS
+21	25	AF‐1	structure_element	Tamoxifen depends on AF‐1 for its cell‐specific activity (Sakamoto et al, 2002); therefore, we asked whether cell‐specific signaling observed here is due to a similar dependence on AF‐1 for activity (Fig EV1).	RESULTS
+181	185	AF‐1	structure_element	Tamoxifen depends on AF‐1 for its cell‐specific activity (Sakamoto et al, 2002); therefore, we asked whether cell‐specific signaling observed here is due to a similar dependence on AF‐1 for activity (Fig EV1).	RESULTS
+35	59	average L‐Luc activities	evidence	To test this idea, we compared the average L‐Luc activities of each scaffold in HepG2 cells co‐transfected with wild‐type ERα or with ERα lacking the AB domain (Figs 1B and EV1).	RESULTS
+92	106	co‐transfected	experimental_method	To test this idea, we compared the average L‐Luc activities of each scaffold in HepG2 cells co‐transfected with wild‐type ERα or with ERα lacking the AB domain (Figs 1B and EV1).	RESULTS
+112	121	wild‐type	protein_state	To test this idea, we compared the average L‐Luc activities of each scaffold in HepG2 cells co‐transfected with wild‐type ERα or with ERα lacking the AB domain (Figs 1B and EV1).	RESULTS
+122	125	ERα	protein	To test this idea, we compared the average L‐Luc activities of each scaffold in HepG2 cells co‐transfected with wild‐type ERα or with ERα lacking the AB domain (Figs 1B and EV1).	RESULTS
+134	137	ERα	protein	To test this idea, we compared the average L‐Luc activities of each scaffold in HepG2 cells co‐transfected with wild‐type ERα or with ERα lacking the AB domain (Figs 1B and EV1).	RESULTS
+138	149	lacking the	protein_state	To test this idea, we compared the average L‐Luc activities of each scaffold in HepG2 cells co‐transfected with wild‐type ERα or with ERα lacking the AB domain (Figs 1B and EV1).	RESULTS
+150	152	AB	structure_element	To test this idea, we compared the average L‐Luc activities of each scaffold in HepG2 cells co‐transfected with wild‐type ERα or with ERα lacking the AB domain (Figs 1B and EV1).	RESULTS
+6	8	E2	chemical	While E2 showed similar L‐Luc ERα‐WT and ERα‐ΔAB activities, tamoxifen showed complete loss of activity without the AB domain (Fig EV1B).	RESULTS
+24	29	L‐Luc	experimental_method	While E2 showed similar L‐Luc ERα‐WT and ERα‐ΔAB activities, tamoxifen showed complete loss of activity without the AB domain (Fig EV1B).	RESULTS
+30	33	ERα	protein	While E2 showed similar L‐Luc ERα‐WT and ERα‐ΔAB activities, tamoxifen showed complete loss of activity without the AB domain (Fig EV1B).	RESULTS
+34	36	WT	protein_state	While E2 showed similar L‐Luc ERα‐WT and ERα‐ΔAB activities, tamoxifen showed complete loss of activity without the AB domain (Fig EV1B).	RESULTS
+41	48	ERα‐ΔAB	mutant	While E2 showed similar L‐Luc ERα‐WT and ERα‐ΔAB activities, tamoxifen showed complete loss of activity without the AB domain (Fig EV1B).	RESULTS
+61	70	tamoxifen	chemical	While E2 showed similar L‐Luc ERα‐WT and ERα‐ΔAB activities, tamoxifen showed complete loss of activity without the AB domain (Fig EV1B).	RESULTS
+104	111	without	protein_state	While E2 showed similar L‐Luc ERα‐WT and ERα‐ΔAB activities, tamoxifen showed complete loss of activity without the AB domain (Fig EV1B).	RESULTS
+116	118	AB	structure_element	While E2 showed similar L‐Luc ERα‐WT and ERα‐ΔAB activities, tamoxifen showed complete loss of activity without the AB domain (Fig EV1B).	RESULTS
+0	11	Deletion of	experimental_method	Deletion of the AB domain significantly reduced the average L‐Luc activities of 14 scaffolds (Student's t‐test, P ≤ 0.05) (Fig 3B).	RESULTS
+16	18	AB	structure_element	Deletion of the AB domain significantly reduced the average L‐Luc activities of 14 scaffolds (Student's t‐test, P ≤ 0.05) (Fig 3B).	RESULTS
+52	76	average L‐Luc activities	evidence	Deletion of the AB domain significantly reduced the average L‐Luc activities of 14 scaffolds (Student's t‐test, P ≤ 0.05) (Fig 3B).	RESULTS
+94	110	Student's t‐test	experimental_method	Deletion of the AB domain significantly reduced the average L‐Luc activities of 14 scaffolds (Student's t‐test, P ≤ 0.05) (Fig 3B).	RESULTS
+112	113	P	evidence	Deletion of the AB domain significantly reduced the average L‐Luc activities of 14 scaffolds (Student's t‐test, P ≤ 0.05) (Fig 3B).	RESULTS
+7	11	AF‐1	structure_element	These “AF‐1‐sensitive” activities were exhibited by both direct and indirect modulators, and were not limited to scaffolds that showed cell‐specific signaling (Fig 3A and B).	RESULTS
+22	26	AF‐1	structure_element	Thus, the strength of AF‐1 signaling does not determine cell‐specific signaling.	RESULTS
+48	51	ERα	protein	Identifying cell‐specific signaling clusters in ERα ligand classes	RESULTS
+53	86	Pearson's correlation coefficient	evidence	For each ligand class or scaffold, we calculated the Pearson's correlation coefficient, r, for pairwise comparison of activity profiles in breast (GREB1), liver (L‐Luc), and endometrial cells (E‐Luc).	RESULTS
+88	89	r	evidence	For each ligand class or scaffold, we calculated the Pearson's correlation coefficient, r, for pairwise comparison of activity profiles in breast (GREB1), liver (L‐Luc), and endometrial cells (E‐Luc).	RESULTS
+147	152	GREB1	protein	For each ligand class or scaffold, we calculated the Pearson's correlation coefficient, r, for pairwise comparison of activity profiles in breast (GREB1), liver (L‐Luc), and endometrial cells (E‐Luc).	RESULTS
+162	167	L‐Luc	experimental_method	For each ligand class or scaffold, we calculated the Pearson's correlation coefficient, r, for pairwise comparison of activity profiles in breast (GREB1), liver (L‐Luc), and endometrial cells (E‐Luc).	RESULTS
+193	198	E‐Luc	experimental_method	For each ligand class or scaffold, we calculated the Pearson's correlation coefficient, r, for pairwise comparison of activity profiles in breast (GREB1), liver (L‐Luc), and endometrial cells (E‐Luc).	RESULTS
+13	14	r	evidence	The value of r ranges from −1 to 1, and it defines the extent to which the data fit a straight line when compounds show similar agonist/antagonist activity profiles between cell types (Fig EV3A).	RESULTS
+23	51	coefficient of determination	evidence	We also calculated the coefficient of determination, r 2, which describes the percentage of variance in a dependent variable such as proliferation that can be predicted by an independent variable such as GREB1 expression.	RESULTS
+53	56	r 2	evidence	We also calculated the coefficient of determination, r 2, which describes the percentage of variance in a dependent variable such as proliferation that can be predicted by an independent variable such as GREB1 expression.	RESULTS
+204	209	GREB1	protein	We also calculated the coefficient of determination, r 2, which describes the percentage of variance in a dependent variable such as proliferation that can be predicted by an independent variable such as GREB1 expression.	RESULTS
+32	35	r 2	evidence	We present both calculations as r 2 to readily compare signaling specificities using a heat map on which the red–yellow palette indicates significant positive correlations (P ≤ 0.05, F‐test for nonzero slope), while the blue palette denotes negative correlations (Fig 3C–F).	RESULTS
+173	174	P	evidence	We present both calculations as r 2 to readily compare signaling specificities using a heat map on which the red–yellow palette indicates significant positive correlations (P ≤ 0.05, F‐test for nonzero slope), while the blue palette denotes negative correlations (Fig 3C–F).	RESULTS
+183	189	F‐test	experimental_method	We present both calculations as r 2 to readily compare signaling specificities using a heat map on which the red–yellow palette indicates significant positive correlations (P ≤ 0.05, F‐test for nonzero slope), while the blue palette denotes negative correlations (Fig 3C–F).	RESULTS
+18	26	OBHS‐BSC	chemical	The side chain of OBHS‐BSC analogs induces cell‐specific signaling	FIG
+24	28	OBHS	chemical	Correlation analysis of OBHS versus OBHS‐BSC activity across cell types.	FIG
+36	44	OBHS‐BSC	chemical	Correlation analysis of OBHS versus OBHS‐BSC activity across cell types.	FIG
+24	29	L‐Luc	experimental_method	Correlation analysis of L‐Luc ERα‐ΔAB activity versus endogenous ERα activity of OBHS analogs.	FIG
+30	37	ERα‐ΔAB	mutant	Correlation analysis of L‐Luc ERα‐ΔAB activity versus endogenous ERα activity of OBHS analogs.	FIG
+65	68	ERα	protein	Correlation analysis of L‐Luc ERα‐ΔAB activity versus endogenous ERα activity of OBHS analogs.	FIG
+81	85	OBHS	chemical	Correlation analysis of L‐Luc ERα‐ΔAB activity versus endogenous ERα activity of OBHS analogs.	FIG
+14	19	L‐Luc	experimental_method	In panel (D), L‐Luc ERα‐WT activity from panel (B) is shown for comparison.	FIG
+20	23	ERα	protein	In panel (D), L‐Luc ERα‐WT activity from panel (B) is shown for comparison.	FIG
+24	26	WT	protein_state	In panel (D), L‐Luc ERα‐WT activity from panel (B) is shown for comparison.	FIG
+24	29	L‐Luc	experimental_method	Correlation analysis of L‐Luc ERα‐ΔF activity versus endogenous ERα activities of OBHS analogs.	FIG
+30	36	ERα‐ΔF	mutant	Correlation analysis of L‐Luc ERα‐ΔF activity versus endogenous ERα activities of OBHS analogs.	FIG
+64	67	ERα	protein	Correlation analysis of L‐Luc ERα‐ΔF activity versus endogenous ERα activities of OBHS analogs.	FIG
+82	86	OBHS	chemical	Correlation analysis of L‐Luc ERα‐ΔF activity versus endogenous ERα activities of OBHS analogs.	FIG
+56	63	NCOA2/3	protein	Correlation analysis of MCF‐7 cell proliferation versus NCOA2/3 recruitment or GREB1 levels observed in response to (G) OBHS‐N and (H) OBHS‐BSC analogs.	FIG
+79	84	GREB1	protein	Correlation analysis of MCF‐7 cell proliferation versus NCOA2/3 recruitment or GREB1 levels observed in response to (G) OBHS‐N and (H) OBHS‐BSC analogs.	FIG
+120	126	OBHS‐N	chemical	Correlation analysis of MCF‐7 cell proliferation versus NCOA2/3 recruitment or GREB1 levels observed in response to (G) OBHS‐N and (H) OBHS‐BSC analogs.	FIG
+135	143	OBHS‐BSC	chemical	Correlation analysis of MCF‐7 cell proliferation versus NCOA2/3 recruitment or GREB1 levels observed in response to (G) OBHS‐N and (H) OBHS‐BSC analogs.	FIG
+53	58	GREB1	protein	Scaffolds in cluster 1 exhibited strongly correlated GREB1 levels, E‐Luc and L‐Luc activity profiles across the three cell types (Fig 3C lanes 1–4), suggesting these ligands use similar ERα signaling pathways in the breast, endometrial, and liver cell types.	RESULTS
+67	72	E‐Luc	experimental_method	Scaffolds in cluster 1 exhibited strongly correlated GREB1 levels, E‐Luc and L‐Luc activity profiles across the three cell types (Fig 3C lanes 1–4), suggesting these ligands use similar ERα signaling pathways in the breast, endometrial, and liver cell types.	RESULTS
+77	82	L‐Luc	experimental_method	Scaffolds in cluster 1 exhibited strongly correlated GREB1 levels, E‐Luc and L‐Luc activity profiles across the three cell types (Fig 3C lanes 1–4), suggesting these ligands use similar ERα signaling pathways in the breast, endometrial, and liver cell types.	RESULTS
+186	189	ERα	protein	Scaffolds in cluster 1 exhibited strongly correlated GREB1 levels, E‐Luc and L‐Luc activity profiles across the three cell types (Fig 3C lanes 1–4), suggesting these ligands use similar ERα signaling pathways in the breast, endometrial, and liver cell types.	RESULTS
+22	27	WAY‐C	chemical	This cluster includes WAY‐C, OBHS, OBHS‐N, and triaryl‐ethylene analogs, all of which are indirect modulators.	RESULTS
+29	33	OBHS	chemical	This cluster includes WAY‐C, OBHS, OBHS‐N, and triaryl‐ethylene analogs, all of which are indirect modulators.	RESULTS
+35	41	OBHS‐N	chemical	This cluster includes WAY‐C, OBHS, OBHS‐N, and triaryl‐ethylene analogs, all of which are indirect modulators.	RESULTS
+47	63	triaryl‐ethylene	chemical	This cluster includes WAY‐C, OBHS, OBHS‐N, and triaryl‐ethylene analogs, all of which are indirect modulators.	RESULTS
+56	70	cyclofenil‐ASC	chemical	This cluster includes two classes of direct modulators (cyclofenil‐ASC and WAY dimer), and six classes of indirect modulators (2,5‐DTP, 3,4‐DTP, S‐OBHS‐2 and S‐OBHS‐3, furan, and WAY‐D).	RESULTS
+75	84	WAY dimer	chemical	This cluster includes two classes of direct modulators (cyclofenil‐ASC and WAY dimer), and six classes of indirect modulators (2,5‐DTP, 3,4‐DTP, S‐OBHS‐2 and S‐OBHS‐3, furan, and WAY‐D).	RESULTS
+127	134	2,5‐DTP	chemical	This cluster includes two classes of direct modulators (cyclofenil‐ASC and WAY dimer), and six classes of indirect modulators (2,5‐DTP, 3,4‐DTP, S‐OBHS‐2 and S‐OBHS‐3, furan, and WAY‐D).	RESULTS
+136	143	3,4‐DTP	chemical	This cluster includes two classes of direct modulators (cyclofenil‐ASC and WAY dimer), and six classes of indirect modulators (2,5‐DTP, 3,4‐DTP, S‐OBHS‐2 and S‐OBHS‐3, furan, and WAY‐D).	RESULTS
+145	153	S‐OBHS‐2	chemical	This cluster includes two classes of direct modulators (cyclofenil‐ASC and WAY dimer), and six classes of indirect modulators (2,5‐DTP, 3,4‐DTP, S‐OBHS‐2 and S‐OBHS‐3, furan, and WAY‐D).	RESULTS
+158	166	S‐OBHS‐3	chemical	This cluster includes two classes of direct modulators (cyclofenil‐ASC and WAY dimer), and six classes of indirect modulators (2,5‐DTP, 3,4‐DTP, S‐OBHS‐2 and S‐OBHS‐3, furan, and WAY‐D).	RESULTS
+168	173	furan	chemical	This cluster includes two classes of direct modulators (cyclofenil‐ASC and WAY dimer), and six classes of indirect modulators (2,5‐DTP, 3,4‐DTP, S‐OBHS‐2 and S‐OBHS‐3, furan, and WAY‐D).	RESULTS
+179	184	WAY‐D	chemical	This cluster includes two classes of direct modulators (cyclofenil‐ASC and WAY dimer), and six classes of indirect modulators (2,5‐DTP, 3,4‐DTP, S‐OBHS‐2 and S‐OBHS‐3, furan, and WAY‐D).	RESULTS
+13	20	3,4‐DTP	chemical	For example, 3,4‐DTP, furan, and S‐OBHS‐2 drove positively correlated GREB1 levels and E‐Luc but not L‐Luc ERα‐WT activity (Fig 3C lanes 5–7).	RESULTS
+22	27	furan	chemical	For example, 3,4‐DTP, furan, and S‐OBHS‐2 drove positively correlated GREB1 levels and E‐Luc but not L‐Luc ERα‐WT activity (Fig 3C lanes 5–7).	RESULTS
+33	41	S‐OBHS‐2	chemical	For example, 3,4‐DTP, furan, and S‐OBHS‐2 drove positively correlated GREB1 levels and E‐Luc but not L‐Luc ERα‐WT activity (Fig 3C lanes 5–7).	RESULTS
+70	75	GREB1	protein	For example, 3,4‐DTP, furan, and S‐OBHS‐2 drove positively correlated GREB1 levels and E‐Luc but not L‐Luc ERα‐WT activity (Fig 3C lanes 5–7).	RESULTS
+87	92	E‐Luc	experimental_method	For example, 3,4‐DTP, furan, and S‐OBHS‐2 drove positively correlated GREB1 levels and E‐Luc but not L‐Luc ERα‐WT activity (Fig 3C lanes 5–7).	RESULTS
+101	106	L‐Luc	experimental_method	For example, 3,4‐DTP, furan, and S‐OBHS‐2 drove positively correlated GREB1 levels and E‐Luc but not L‐Luc ERα‐WT activity (Fig 3C lanes 5–7).	RESULTS
+107	110	ERα	protein	For example, 3,4‐DTP, furan, and S‐OBHS‐2 drove positively correlated GREB1 levels and E‐Luc but not L‐Luc ERα‐WT activity (Fig 3C lanes 5–7).	RESULTS
+111	113	WT	protein_state	For example, 3,4‐DTP, furan, and S‐OBHS‐2 drove positively correlated GREB1 levels and E‐Luc but not L‐Luc ERα‐WT activity (Fig 3C lanes 5–7).	RESULTS
+13	22	WAY dimer	chemical	In contrast, WAY dimer and WAY‐D analogs drove positively correlated GREB1 levels and L‐Luc ERα‐WT but not E‐Luc activity (Fig 3C lanes 8 and 9).	RESULTS
+27	32	WAY‐D	chemical	In contrast, WAY dimer and WAY‐D analogs drove positively correlated GREB1 levels and L‐Luc ERα‐WT but not E‐Luc activity (Fig 3C lanes 8 and 9).	RESULTS
+69	74	GREB1	protein	In contrast, WAY dimer and WAY‐D analogs drove positively correlated GREB1 levels and L‐Luc ERα‐WT but not E‐Luc activity (Fig 3C lanes 8 and 9).	RESULTS
+86	91	L‐Luc	experimental_method	In contrast, WAY dimer and WAY‐D analogs drove positively correlated GREB1 levels and L‐Luc ERα‐WT but not E‐Luc activity (Fig 3C lanes 8 and 9).	RESULTS
+92	95	ERα	protein	In contrast, WAY dimer and WAY‐D analogs drove positively correlated GREB1 levels and L‐Luc ERα‐WT but not E‐Luc activity (Fig 3C lanes 8 and 9).	RESULTS
+96	98	WT	protein_state	In contrast, WAY dimer and WAY‐D analogs drove positively correlated GREB1 levels and L‐Luc ERα‐WT but not E‐Luc activity (Fig 3C lanes 8 and 9).	RESULTS
+107	112	E‐Luc	experimental_method	In contrast, WAY dimer and WAY‐D analogs drove positively correlated GREB1 levels and L‐Luc ERα‐WT but not E‐Luc activity (Fig 3C lanes 8 and 9).	RESULTS
+54	62	OBHS‐ASC	chemical	This cluster includes two direct modulator scaffolds (OBHS‐ASC and OBHS‐BSC), and five indirect modulator scaffolds (A‐CD, cyclofenil, 3,4‐DTPD, imine, and imidazopyridine).	RESULTS
+67	75	OBHS‐BSC	chemical	This cluster includes two direct modulator scaffolds (OBHS‐ASC and OBHS‐BSC), and five indirect modulator scaffolds (A‐CD, cyclofenil, 3,4‐DTPD, imine, and imidazopyridine).	RESULTS
+117	121	A‐CD	chemical	This cluster includes two direct modulator scaffolds (OBHS‐ASC and OBHS‐BSC), and five indirect modulator scaffolds (A‐CD, cyclofenil, 3,4‐DTPD, imine, and imidazopyridine).	RESULTS
+123	133	cyclofenil	chemical	This cluster includes two direct modulator scaffolds (OBHS‐ASC and OBHS‐BSC), and five indirect modulator scaffolds (A‐CD, cyclofenil, 3,4‐DTPD, imine, and imidazopyridine).	RESULTS
+135	143	3,4‐DTPD	chemical	This cluster includes two direct modulator scaffolds (OBHS‐ASC and OBHS‐BSC), and five indirect modulator scaffolds (A‐CD, cyclofenil, 3,4‐DTPD, imine, and imidazopyridine).	RESULTS
+145	150	imine	chemical	This cluster includes two direct modulator scaffolds (OBHS‐ASC and OBHS‐BSC), and five indirect modulator scaffolds (A‐CD, cyclofenil, 3,4‐DTPD, imine, and imidazopyridine).	RESULTS
+156	171	imidazopyridine	chemical	This cluster includes two direct modulator scaffolds (OBHS‐ASC and OBHS‐BSC), and five indirect modulator scaffolds (A‐CD, cyclofenil, 3,4‐DTPD, imine, and imidazopyridine).	RESULTS
+68	71	ERα	protein	These results suggest that addition of an extended side chain to an ERα ligand scaffold is sufficient to induce cell‐specific signaling, where the relative activity profiles of the individual ligands change between cell types.	RESULTS
+82	86	OBHS	chemical	This is demonstrated by directly comparing the signaling specificities of matched OBHS (indirect modulator, cluster 1) and OBHS‐BSC analogs (direct modulator, cluster 3), which differ only in the basic side chain (Fig 2E).	RESULTS
+123	131	OBHS‐BSC	chemical	This is demonstrated by directly comparing the signaling specificities of matched OBHS (indirect modulator, cluster 1) and OBHS‐BSC analogs (direct modulator, cluster 3), which differ only in the basic side chain (Fig 2E).	RESULTS
+18	22	OBHS	chemical	The activities of OBHS analogs were positively correlated across the three cell types, but the side chain of OBHS‐BSC analogs was sufficient to abolish these correlations (Figs 3C lanes 1 and 19, and EV3A–C).	RESULTS
+109	117	OBHS‐BSC	chemical	The activities of OBHS analogs were positively correlated across the three cell types, but the side chain of OBHS‐BSC analogs was sufficient to abolish these correlations (Figs 3C lanes 1 and 19, and EV3A–C).	RESULTS
+43	46	ERα	protein	Thus, examining the correlated patterns of ERα activity within each scaffold demonstrates that an extended side chain is not required for cell‐specific signaling.	RESULTS
+39	43	AF‐1	structure_element	Modulation of signaling specificity by AF‐1	RESULTS
+24	28	AF‐1	structure_element	To evaluate the role of AF‐1 and the F domain in ERα signaling specificity, we compared activity of truncated ERα constructs in HepG2 liver cells with endogenous ERα activity in the other cell types.	RESULTS
+37	38	F	structure_element	To evaluate the role of AF‐1 and the F domain in ERα signaling specificity, we compared activity of truncated ERα constructs in HepG2 liver cells with endogenous ERα activity in the other cell types.	RESULTS
+49	52	ERα	protein	To evaluate the role of AF‐1 and the F domain in ERα signaling specificity, we compared activity of truncated ERα constructs in HepG2 liver cells with endogenous ERα activity in the other cell types.	RESULTS
+110	113	ERα	protein	To evaluate the role of AF‐1 and the F domain in ERα signaling specificity, we compared activity of truncated ERα constructs in HepG2 liver cells with endogenous ERα activity in the other cell types.	RESULTS
+162	165	ERα	protein	To evaluate the role of AF‐1 and the F domain in ERα signaling specificity, we compared activity of truncated ERα constructs in HepG2 liver cells with endogenous ERα activity in the other cell types.	RESULTS
+37	42	L‐Luc	experimental_method	The positive correlation between the L‐Luc and E‐Luc activities or GREB1 levels induced by scaffolds in cluster 1 was generally retained without the AB domain, or the F domain (Fig 3D lanes 1–4).	RESULTS
+47	52	E‐Luc	experimental_method	The positive correlation between the L‐Luc and E‐Luc activities or GREB1 levels induced by scaffolds in cluster 1 was generally retained without the AB domain, or the F domain (Fig 3D lanes 1–4).	RESULTS
+67	72	GREB1	protein	The positive correlation between the L‐Luc and E‐Luc activities or GREB1 levels induced by scaffolds in cluster 1 was generally retained without the AB domain, or the F domain (Fig 3D lanes 1–4).	RESULTS
+149	151	AB	structure_element	The positive correlation between the L‐Luc and E‐Luc activities or GREB1 levels induced by scaffolds in cluster 1 was generally retained without the AB domain, or the F domain (Fig 3D lanes 1–4).	RESULTS
+167	168	F	structure_element	The positive correlation between the L‐Luc and E‐Luc activities or GREB1 levels induced by scaffolds in cluster 1 was generally retained without the AB domain, or the F domain (Fig 3D lanes 1–4).	RESULTS
+110	114	AF‐1	structure_element	This demonstrates that the signaling specificities underlying these positive correlations are not modified by AF‐1.	RESULTS
+0	4	OBHS	chemical	OBHS analogs showed an average L‐Luc ERα‐ΔAB activity of 3.2% ± 3 (mean + SEM) relative to E2.	RESULTS
+31	36	L‐Luc	experimental_method	OBHS analogs showed an average L‐Luc ERα‐ΔAB activity of 3.2% ± 3 (mean + SEM) relative to E2.	RESULTS
+37	44	ERα‐ΔAB	mutant	OBHS analogs showed an average L‐Luc ERα‐ΔAB activity of 3.2% ± 3 (mean + SEM) relative to E2.	RESULTS
+91	93	E2	chemical	OBHS analogs showed an average L‐Luc ERα‐ΔAB activity of 3.2% ± 3 (mean + SEM) relative to E2.	RESULTS
+62	67	L‐Luc	experimental_method	Despite this nearly complete lack of activity, the pattern of L‐Luc ERα‐ΔAB activity was still highly correlated with the E‐Luc activity and GREB1 expression (Fig EV3D and E), demonstrating that very small AF‐2 activities can be amplified by AF‐1 to produce robust signals.	RESULTS
+68	75	ERα‐ΔAB	mutant	Despite this nearly complete lack of activity, the pattern of L‐Luc ERα‐ΔAB activity was still highly correlated with the E‐Luc activity and GREB1 expression (Fig EV3D and E), demonstrating that very small AF‐2 activities can be amplified by AF‐1 to produce robust signals.	RESULTS
+122	127	E‐Luc	experimental_method	Despite this nearly complete lack of activity, the pattern of L‐Luc ERα‐ΔAB activity was still highly correlated with the E‐Luc activity and GREB1 expression (Fig EV3D and E), demonstrating that very small AF‐2 activities can be amplified by AF‐1 to produce robust signals.	RESULTS
+141	146	GREB1	protein	Despite this nearly complete lack of activity, the pattern of L‐Luc ERα‐ΔAB activity was still highly correlated with the E‐Luc activity and GREB1 expression (Fig EV3D and E), demonstrating that very small AF‐2 activities can be amplified by AF‐1 to produce robust signals.	RESULTS
+206	210	AF‐2	structure_element	Despite this nearly complete lack of activity, the pattern of L‐Luc ERα‐ΔAB activity was still highly correlated with the E‐Luc activity and GREB1 expression (Fig EV3D and E), demonstrating that very small AF‐2 activities can be amplified by AF‐1 to produce robust signals.	RESULTS
+242	246	AF‐1	structure_element	Despite this nearly complete lack of activity, the pattern of L‐Luc ERα‐ΔAB activity was still highly correlated with the E‐Luc activity and GREB1 expression (Fig EV3D and E), demonstrating that very small AF‐2 activities can be amplified by AF‐1 to produce robust signals.	RESULTS
+11	22	deletion of	experimental_method	Similarly, deletion of the F domain did not abolish correlations between the L‐Luc and E‐Luc or GREB1 levels induced by OBHS analogs (Fig EV3F).	RESULTS
+27	28	F	structure_element	Similarly, deletion of the F domain did not abolish correlations between the L‐Luc and E‐Luc or GREB1 levels induced by OBHS analogs (Fig EV3F).	RESULTS
+77	82	L‐Luc	experimental_method	Similarly, deletion of the F domain did not abolish correlations between the L‐Luc and E‐Luc or GREB1 levels induced by OBHS analogs (Fig EV3F).	RESULTS
+87	92	E‐Luc	experimental_method	Similarly, deletion of the F domain did not abolish correlations between the L‐Luc and E‐Luc or GREB1 levels induced by OBHS analogs (Fig EV3F).	RESULTS
+96	101	GREB1	protein	Similarly, deletion of the F domain did not abolish correlations between the L‐Luc and E‐Luc or GREB1 levels induced by OBHS analogs (Fig EV3F).	RESULTS
+120	124	OBHS	chemical	Similarly, deletion of the F domain did not abolish correlations between the L‐Luc and E‐Luc or GREB1 levels induced by OBHS analogs (Fig EV3F).	RESULTS
+49	58	wild‐type	protein_state	These similar patterns of ligand activity in the wild‐type and deletion mutants suggest that AF‐1 and the F domain purely amplify the AF‐2 activities of ligands in cluster 1.	RESULTS
+72	79	mutants	protein_state	These similar patterns of ligand activity in the wild‐type and deletion mutants suggest that AF‐1 and the F domain purely amplify the AF‐2 activities of ligands in cluster 1.	RESULTS
+93	97	AF‐1	structure_element	These similar patterns of ligand activity in the wild‐type and deletion mutants suggest that AF‐1 and the F domain purely amplify the AF‐2 activities of ligands in cluster 1.	RESULTS
+106	107	F	structure_element	These similar patterns of ligand activity in the wild‐type and deletion mutants suggest that AF‐1 and the F domain purely amplify the AF‐2 activities of ligands in cluster 1.	RESULTS
+134	138	AF‐2	structure_element	These similar patterns of ligand activity in the wild‐type and deletion mutants suggest that AF‐1 and the F domain purely amplify the AF‐2 activities of ligands in cluster 1.	RESULTS
+13	17	AF‐1	structure_element	In contrast, AF‐1 was a determinant of signaling specificity for scaffolds in cluster 2.	RESULTS
+0	11	Deletion of	experimental_method	Deletion of the AB or F domain altered correlations for six of the eight scaffolds in this cluster (2,5‐DTP, 3,4‐DTP, S‐OBHS‐3, WAY‐D, WAY dimer, and cyclofenil‐ASC) (Fig 3D lanes 5–12).	RESULTS
+16	18	AB	structure_element	Deletion of the AB or F domain altered correlations for six of the eight scaffolds in this cluster (2,5‐DTP, 3,4‐DTP, S‐OBHS‐3, WAY‐D, WAY dimer, and cyclofenil‐ASC) (Fig 3D lanes 5–12).	RESULTS
+22	23	F	structure_element	Deletion of the AB or F domain altered correlations for six of the eight scaffolds in this cluster (2,5‐DTP, 3,4‐DTP, S‐OBHS‐3, WAY‐D, WAY dimer, and cyclofenil‐ASC) (Fig 3D lanes 5–12).	RESULTS
+100	107	2,5‐DTP	chemical	Deletion of the AB or F domain altered correlations for six of the eight scaffolds in this cluster (2,5‐DTP, 3,4‐DTP, S‐OBHS‐3, WAY‐D, WAY dimer, and cyclofenil‐ASC) (Fig 3D lanes 5–12).	RESULTS
+109	116	3,4‐DTP	chemical	Deletion of the AB or F domain altered correlations for six of the eight scaffolds in this cluster (2,5‐DTP, 3,4‐DTP, S‐OBHS‐3, WAY‐D, WAY dimer, and cyclofenil‐ASC) (Fig 3D lanes 5–12).	RESULTS
+118	126	S‐OBHS‐3	chemical	Deletion of the AB or F domain altered correlations for six of the eight scaffolds in this cluster (2,5‐DTP, 3,4‐DTP, S‐OBHS‐3, WAY‐D, WAY dimer, and cyclofenil‐ASC) (Fig 3D lanes 5–12).	RESULTS
+128	133	WAY‐D	chemical	Deletion of the AB or F domain altered correlations for six of the eight scaffolds in this cluster (2,5‐DTP, 3,4‐DTP, S‐OBHS‐3, WAY‐D, WAY dimer, and cyclofenil‐ASC) (Fig 3D lanes 5–12).	RESULTS
+135	144	WAY dimer	chemical	Deletion of the AB or F domain altered correlations for six of the eight scaffolds in this cluster (2,5‐DTP, 3,4‐DTP, S‐OBHS‐3, WAY‐D, WAY dimer, and cyclofenil‐ASC) (Fig 3D lanes 5–12).	RESULTS
+150	164	cyclofenil‐ASC	chemical	Deletion of the AB or F domain altered correlations for six of the eight scaffolds in this cluster (2,5‐DTP, 3,4‐DTP, S‐OBHS‐3, WAY‐D, WAY dimer, and cyclofenil‐ASC) (Fig 3D lanes 5–12).	RESULTS
+61	83	deletion mutant assays	experimental_method	Comparing Fig 3C and D, the + and − signs indicate where the deletion mutant assays led to a gain or loss of statically significant correlation, respectively.	RESULTS
+20	24	AF‐1	structure_element	Thus, in cluster 2, AF‐1 substantially modulated the specificity of ligands with cell‐specific activity (Fig 3D lanes 5–12).	RESULTS
+60	64	AF‐1	structure_element	For ligands in cluster 3, we could not eliminate a role for AF‐1 in determining signaling specificity, since this cluster lacked positively correlated activity profiles (Fig 3C), and deletion of the AB or F domain rarely induced such correlations (Fig 3D), except for A‐CD and OBHS‐ASC analogs, where deletion of the AB domain or F domain led to positive correlations with E‐Luc activity and/or GREB1 levels (Fig 3D lanes 13 and 18).	RESULTS
+183	194	deletion of	experimental_method	For ligands in cluster 3, we could not eliminate a role for AF‐1 in determining signaling specificity, since this cluster lacked positively correlated activity profiles (Fig 3C), and deletion of the AB or F domain rarely induced such correlations (Fig 3D), except for A‐CD and OBHS‐ASC analogs, where deletion of the AB domain or F domain led to positive correlations with E‐Luc activity and/or GREB1 levels (Fig 3D lanes 13 and 18).	RESULTS
+199	201	AB	structure_element	For ligands in cluster 3, we could not eliminate a role for AF‐1 in determining signaling specificity, since this cluster lacked positively correlated activity profiles (Fig 3C), and deletion of the AB or F domain rarely induced such correlations (Fig 3D), except for A‐CD and OBHS‐ASC analogs, where deletion of the AB domain or F domain led to positive correlations with E‐Luc activity and/or GREB1 levels (Fig 3D lanes 13 and 18).	RESULTS
+205	206	F	structure_element	For ligands in cluster 3, we could not eliminate a role for AF‐1 in determining signaling specificity, since this cluster lacked positively correlated activity profiles (Fig 3C), and deletion of the AB or F domain rarely induced such correlations (Fig 3D), except for A‐CD and OBHS‐ASC analogs, where deletion of the AB domain or F domain led to positive correlations with E‐Luc activity and/or GREB1 levels (Fig 3D lanes 13 and 18).	RESULTS
+268	272	A‐CD	chemical	For ligands in cluster 3, we could not eliminate a role for AF‐1 in determining signaling specificity, since this cluster lacked positively correlated activity profiles (Fig 3C), and deletion of the AB or F domain rarely induced such correlations (Fig 3D), except for A‐CD and OBHS‐ASC analogs, where deletion of the AB domain or F domain led to positive correlations with E‐Luc activity and/or GREB1 levels (Fig 3D lanes 13 and 18).	RESULTS
+277	285	OBHS‐ASC	chemical	For ligands in cluster 3, we could not eliminate a role for AF‐1 in determining signaling specificity, since this cluster lacked positively correlated activity profiles (Fig 3C), and deletion of the AB or F domain rarely induced such correlations (Fig 3D), except for A‐CD and OBHS‐ASC analogs, where deletion of the AB domain or F domain led to positive correlations with E‐Luc activity and/or GREB1 levels (Fig 3D lanes 13 and 18).	RESULTS
+301	312	deletion of	experimental_method	For ligands in cluster 3, we could not eliminate a role for AF‐1 in determining signaling specificity, since this cluster lacked positively correlated activity profiles (Fig 3C), and deletion of the AB or F domain rarely induced such correlations (Fig 3D), except for A‐CD and OBHS‐ASC analogs, where deletion of the AB domain or F domain led to positive correlations with E‐Luc activity and/or GREB1 levels (Fig 3D lanes 13 and 18).	RESULTS
+317	319	AB	structure_element	For ligands in cluster 3, we could not eliminate a role for AF‐1 in determining signaling specificity, since this cluster lacked positively correlated activity profiles (Fig 3C), and deletion of the AB or F domain rarely induced such correlations (Fig 3D), except for A‐CD and OBHS‐ASC analogs, where deletion of the AB domain or F domain led to positive correlations with E‐Luc activity and/or GREB1 levels (Fig 3D lanes 13 and 18).	RESULTS
+330	331	F	structure_element	For ligands in cluster 3, we could not eliminate a role for AF‐1 in determining signaling specificity, since this cluster lacked positively correlated activity profiles (Fig 3C), and deletion of the AB or F domain rarely induced such correlations (Fig 3D), except for A‐CD and OBHS‐ASC analogs, where deletion of the AB domain or F domain led to positive correlations with E‐Luc activity and/or GREB1 levels (Fig 3D lanes 13 and 18).	RESULTS
+373	378	E‐Luc	experimental_method	For ligands in cluster 3, we could not eliminate a role for AF‐1 in determining signaling specificity, since this cluster lacked positively correlated activity profiles (Fig 3C), and deletion of the AB or F domain rarely induced such correlations (Fig 3D), except for A‐CD and OBHS‐ASC analogs, where deletion of the AB domain or F domain led to positive correlations with E‐Luc activity and/or GREB1 levels (Fig 3D lanes 13 and 18).	RESULTS
+395	400	GREB1	protein	For ligands in cluster 3, we could not eliminate a role for AF‐1 in determining signaling specificity, since this cluster lacked positively correlated activity profiles (Fig 3C), and deletion of the AB or F domain rarely induced such correlations (Fig 3D), except for A‐CD and OBHS‐ASC analogs, where deletion of the AB domain or F domain led to positive correlations with E‐Luc activity and/or GREB1 levels (Fig 3D lanes 13 and 18).	RESULTS
+35	39	AF‐1	structure_element	Thus, ligands in cluster 2 rely on AF‐1 for both activity (Fig 3B) and signaling specificity (Fig 3D).	RESULTS
+27	32	GREB1	protein	Ligand‐specific control of GREB1 expression	RESULTS
+65	68	ERα	protein	To determine whether ligand classes control expression of native ERα target genes through the canonical linear signaling pathway, we performed pairwise linear regression analyses using ERα–NCOA1/2/3 interactions in M2H assay as independent predictors of GREB1 expression (the dependent variable) (Figs EV1 and EV2A, F–H).	RESULTS
+143	178	pairwise linear regression analyses	experimental_method	To determine whether ligand classes control expression of native ERα target genes through the canonical linear signaling pathway, we performed pairwise linear regression analyses using ERα–NCOA1/2/3 interactions in M2H assay as independent predictors of GREB1 expression (the dependent variable) (Figs EV1 and EV2A, F–H).	RESULTS
+185	198	ERα–NCOA1/2/3	complex_assembly	To determine whether ligand classes control expression of native ERα target genes through the canonical linear signaling pathway, we performed pairwise linear regression analyses using ERα–NCOA1/2/3 interactions in M2H assay as independent predictors of GREB1 expression (the dependent variable) (Figs EV1 and EV2A, F–H).	RESULTS
+215	224	M2H assay	experimental_method	To determine whether ligand classes control expression of native ERα target genes through the canonical linear signaling pathway, we performed pairwise linear regression analyses using ERα–NCOA1/2/3 interactions in M2H assay as independent predictors of GREB1 expression (the dependent variable) (Figs EV1 and EV2A, F–H).	RESULTS
+254	259	GREB1	protein	To determine whether ligand classes control expression of native ERα target genes through the canonical linear signaling pathway, we performed pairwise linear regression analyses using ERα–NCOA1/2/3 interactions in M2H assay as independent predictors of GREB1 expression (the dependent variable) (Figs EV1 and EV2A, F–H).	RESULTS
+33	38	NCOA1	protein	In cluster 1, the recruitment of NCOA1 and NCOA2 was highest for WAY‐C, followed by triaryl‐ethylene, OBHS‐N, and OBHS series, while for NCOA3, OBHS‐N compounds induced the most recruitment and OBHS ligands were inverse agonists (Fig EV2F–H).	RESULTS
+43	48	NCOA2	protein	In cluster 1, the recruitment of NCOA1 and NCOA2 was highest for WAY‐C, followed by triaryl‐ethylene, OBHS‐N, and OBHS series, while for NCOA3, OBHS‐N compounds induced the most recruitment and OBHS ligands were inverse agonists (Fig EV2F–H).	RESULTS
+65	70	WAY‐C	chemical	In cluster 1, the recruitment of NCOA1 and NCOA2 was highest for WAY‐C, followed by triaryl‐ethylene, OBHS‐N, and OBHS series, while for NCOA3, OBHS‐N compounds induced the most recruitment and OBHS ligands were inverse agonists (Fig EV2F–H).	RESULTS
+84	100	triaryl‐ethylene	chemical	In cluster 1, the recruitment of NCOA1 and NCOA2 was highest for WAY‐C, followed by triaryl‐ethylene, OBHS‐N, and OBHS series, while for NCOA3, OBHS‐N compounds induced the most recruitment and OBHS ligands were inverse agonists (Fig EV2F–H).	RESULTS
+102	108	OBHS‐N	chemical	In cluster 1, the recruitment of NCOA1 and NCOA2 was highest for WAY‐C, followed by triaryl‐ethylene, OBHS‐N, and OBHS series, while for NCOA3, OBHS‐N compounds induced the most recruitment and OBHS ligands were inverse agonists (Fig EV2F–H).	RESULTS
+114	118	OBHS	chemical	In cluster 1, the recruitment of NCOA1 and NCOA2 was highest for WAY‐C, followed by triaryl‐ethylene, OBHS‐N, and OBHS series, while for NCOA3, OBHS‐N compounds induced the most recruitment and OBHS ligands were inverse agonists (Fig EV2F–H).	RESULTS
+137	142	NCOA3	protein	In cluster 1, the recruitment of NCOA1 and NCOA2 was highest for WAY‐C, followed by triaryl‐ethylene, OBHS‐N, and OBHS series, while for NCOA3, OBHS‐N compounds induced the most recruitment and OBHS ligands were inverse agonists (Fig EV2F–H).	RESULTS
+144	150	OBHS‐N	chemical	In cluster 1, the recruitment of NCOA1 and NCOA2 was highest for WAY��C, followed by triaryl‐ethylene, OBHS‐N, and OBHS series, while for NCOA3, OBHS‐N compounds induced the most recruitment and OBHS ligands were inverse agonists (Fig EV2F–H).	RESULTS
+194	198	OBHS	chemical	In cluster 1, the recruitment of NCOA1 and NCOA2 was highest for WAY‐C, followed by triaryl‐ethylene, OBHS‐N, and OBHS series, while for NCOA3, OBHS‐N compounds induced the most recruitment and OBHS ligands were inverse agonists (Fig EV2F–H).	RESULTS
+25	30	GREB1	protein	The average induction of GREB1 by cluster 1 ligands showed greater variance, with a range between ~25 and ~75% for OBHS and a range from full agonist to inverse agonist for the others in cluster 1 (Fig EV2A).	RESULTS
+115	119	OBHS	chemical	The average induction of GREB1 by cluster 1 ligands showed greater variance, with a range between ~25 and ~75% for OBHS and a range from full agonist to inverse agonist for the others in cluster 1 (Fig EV2A).	RESULTS
+0	5	GREB1	protein	GREB1 levels induced by OBHS analogs were determined by recruitment of NCOA1 but not NCOA2/3 (Fig 3E lane 1), suggesting that there may be alternate or preferential use of these coactivators by different classes.	RESULTS
+24	28	OBHS	chemical	GREB1 levels induced by OBHS analogs were determined by recruitment of NCOA1 but not NCOA2/3 (Fig 3E lane 1), suggesting that there may be alternate or preferential use of these coactivators by different classes.	RESULTS
+71	76	NCOA1	protein	GREB1 levels induced by OBHS analogs were determined by recruitment of NCOA1 but not NCOA2/3 (Fig 3E lane 1), suggesting that there may be alternate or preferential use of these coactivators by different classes.	RESULTS
+85	92	NCOA2/3	protein	GREB1 levels induced by OBHS analogs were determined by recruitment of NCOA1 but not NCOA2/3 (Fig 3E lane 1), suggesting that there may be alternate or preferential use of these coactivators by different classes.	RESULTS
+23	32	NCOA1/2/3	protein	However, in cluster 1, NCOA1/2/3 recruitment generally predicted GREB1 levels (Fig 3E lanes 1–4), consistent with the canonical signaling model (Fig 1D).	RESULTS
+65	70	GREB1	protein	However, in cluster 1, NCOA1/2/3 recruitment generally predicted GREB1 levels (Fig 3E lanes 1–4), consistent with the canonical signaling model (Fig 1D).	RESULTS
+22	27	GREB1	protein	For clusters 2 and 3, GREB1 activity was generally not predicted by NCOA1/2/3 recruitment.	RESULTS
+68	77	NCOA1/2/3	protein	For clusters 2 and 3, GREB1 activity was generally not predicted by NCOA1/2/3 recruitment.	RESULTS
+29	38	NCOA1/2/3	protein	Direct modulators showed low NCOA1/2/3 recruitment (Fig EV2F–H), but only OBHS‐ASC analogs had NCOA2 recruitment profiles that predicted a full range of effects on GREB1 levels (Figs 3E lanes 9, 11, 18–19, and EV2A).	RESULTS
+74	82	OBHS‐ASC	chemical	Direct modulators showed low NCOA1/2/3 recruitment (Fig EV2F–H), but only OBHS‐ASC analogs had NCOA2 recruitment profiles that predicted a full range of effects on GREB1 levels (Figs 3E lanes 9, 11, 18–19, and EV2A).	RESULTS
+95	100	NCOA2	protein	Direct modulators showed low NCOA1/2/3 recruitment (Fig EV2F–H), but only OBHS‐ASC analogs had NCOA2 recruitment profiles that predicted a full range of effects on GREB1 levels (Figs 3E lanes 9, 11, 18–19, and EV2A).	RESULTS
+164	169	GREB1	protein	Direct modulators showed low NCOA1/2/3 recruitment (Fig EV2F–H), but only OBHS‐ASC analogs had NCOA2 recruitment profiles that predicted a full range of effects on GREB1 levels (Figs 3E lanes 9, 11, 18–19, and EV2A).	RESULTS
+55	64	NCOA1/2/3	protein	The indirect modulators in clusters 2 and 3 stimulated NCOA1/2/3 recruitment and GREB1 expression with substantial variance (Figs 3A and EV2F–H).	RESULTS
+81	86	GREB1	protein	The indirect modulators in clusters 2 and 3 stimulated NCOA1/2/3 recruitment and GREB1 expression with substantial variance (Figs 3A and EV2F–H).	RESULTS
+24	29	GREB1	protein	However, ligand‐induced GREB1 levels were generally not determined by NCOA1/2/3 recruitment (Fig 3E lanes 5–19), consistent with an alternate causality model (Fig 1E).	RESULTS
+70	79	NCOA1/2/3	protein	However, ligand‐induced GREB1 levels were generally not determined by NCOA1/2/3 recruitment (Fig 3E lanes 5–19), consistent with an alternate causality model (Fig 1E).	RESULTS
+64	72	S‐OBHS‐3	chemical	Out of 11 indirect modulator series in cluster 2 or 3, only the S‐OBHS‐3 class had NCOA1/2/3 recruitment profiles that predicted GREB1 levels (Fig 3E lane 12).	RESULTS
+83	92	NCOA1/2/3	protein	Out of 11 indirect modulator series in cluster 2 or 3, only the S‐OBHS‐3 class had NCOA1/2/3 recruitment profiles that predicted GREB1 levels (Fig 3E lane 12).	RESULTS
+129	134	GREB1	protein	Out of 11 indirect modulator series in cluster 2 or 3, only the S‐OBHS‐3 class had NCOA1/2/3 recruitment profiles that predicted GREB1 levels (Fig 3E lane 12).	RESULTS
+87	92	GREB1	protein	These results suggest that compounds that show cell‐specific signaling do not activate GREB1, or use coactivators other than NCOA1/2/3 to control GREB1 expression (Fig 1E).	RESULTS
+125	134	NCOA1/2/3	protein	These results suggest that compounds that show cell‐specific signaling do not activate GREB1, or use coactivators other than NCOA1/2/3 to control GREB1 expression (Fig 1E).	RESULTS
+146	151	GREB1	protein	These results suggest that compounds that show cell‐specific signaling do not activate GREB1, or use coactivators other than NCOA1/2/3 to control GREB1 expression (Fig 1E).	RESULTS
+103	129	linear regression analyses	experimental_method	To determine mechanisms for ligand‐dependent control of breast cancer cell proliferation, we performed linear regression analyses across the 19 scaffolds using MCF‐7 cell proliferation as the dependent variable, and the other activities as independent variables (Fig 3F).	RESULTS
+14	19	E‐Luc	experimental_method	In cluster 1, E‐Luc and L‐Luc activities, NCOA1/2/3 recruitment, and GREB1 levels generally predicted the proliferative response (Fig 3F lanes 2–4).	RESULTS
+24	29	L‐Luc	experimental_method	In cluster 1, E‐Luc and L‐Luc activities, NCOA1/2/3 recruitment, and GREB1 levels generally predicted the proliferative response (Fig 3F lanes 2–4).	RESULTS
+42	51	NCOA1/2/3	protein	In cluster 1, E‐Luc and L‐Luc activities, NCOA1/2/3 recruitment, and GREB1 levels generally predicted the proliferative response (Fig 3F lanes 2–4).	RESULTS
+69	74	GREB1	protein	In cluster 1, E‐Luc and L‐Luc activities, NCOA1/2/3 recruitment, and GREB1 levels generally predicted the proliferative response (Fig 3F lanes 2–4).	RESULTS
+9	15	OBHS‐N	chemical	With the OBHS‐N compounds, NCOA3 and GREB1 showed near perfect prediction of proliferation (Fig EV3G), with unexplained variance similar to the noise in the assays.	RESULTS
+27	32	NCOA3	protein	With the OBHS‐N compounds, NCOA3 and GREB1 showed near perfect prediction of proliferation (Fig EV3G), with unexplained variance similar to the noise in the assays.	RESULTS
+37	42	GREB1	protein	With the OBHS‐N compounds, NCOA3 and GREB1 showed near perfect prediction of proliferation (Fig EV3G), with unexplained variance similar to the noise in the assays.	RESULTS
+39	43	OBHS	chemical	The lack of significant predictors for OBHS analogs (Fig 3F lane 1) reflects their small range of proliferative effects on MCF‐7 cells (Fig EV2I).	RESULTS
+34	39	GREB1	protein	The significant correlations with GREB1 expression and NCOA1/2/3 recruitment observed in this cluster are consistent with the canonical signaling model (Fig 1D), where NCOA1/2/3 recruitment determines GREB1 expression, which then drives proliferation.	RESULTS
+55	64	NCOA1/2/3	protein	The significant correlations with GREB1 expression and NCOA1/2/3 recruitment observed in this cluster are consistent with the canonical signaling model (Fig 1D), where NCOA1/2/3 recruitment determines GREB1 expression, which then drives proliferation.	RESULTS
+168	177	NCOA1/2/3	protein	The significant correlations with GREB1 expression and NCOA1/2/3 recruitment observed in this cluster are consistent with the canonical signaling model (Fig 1D), where NCOA1/2/3 recruitment determines GREB1 expression, which then drives proliferation.	RESULTS
+201	206	GREB1	protein	The significant correlations with GREB1 expression and NCOA1/2/3 recruitment observed in this cluster are consistent with the canonical signaling model (Fig 1D), where NCOA1/2/3 recruitment determines GREB1 expression, which then drives proliferation.	RESULTS
+90	99	NCOA1/2/3	protein	Despite this phenotypic variance, proliferation was not generally predicted by correlated NCOA1/2/3 recruitment and GREB1 induction (Figs 3F lanes 5–19, and EV3H).	RESULTS
+116	121	GREB1	protein	Despite this phenotypic variance, proliferation was not generally predicted by correlated NCOA1/2/3 recruitment and GREB1 induction (Figs 3F lanes 5–19, and EV3H).	RESULTS
+48	55	2,5‐DTP	chemical	Out of 15 ligand series in these clusters, only 2,5‐DTP analogs induced a proliferative response that was predicted by GREB1 levels, which were not determined by NCOA1/2/3 recruitment (Fig 3E and F lane 10).	RESULTS
+119	124	GREB1	protein	Out of 15 ligand series in these clusters, only 2,5‐DTP analogs induced a proliferative response that was predicted by GREB1 levels, which were not determined by NCOA1/2/3 recruitment (Fig 3E and F lane 10).	RESULTS
+162	171	NCOA1/2/3	protein	Out of 15 ligand series in these clusters, only 2,5‐DTP analogs induced a proliferative response that was predicted by GREB1 levels, which were not determined by NCOA1/2/3 recruitment (Fig 3E and F lane 10).	RESULTS
+0	7	3,4‐DTP	chemical	3,4‐DTP, cyclofenil, 3,4‐DTPD, and imidazopyridine analogs had NCOA1/3 recruitment profiles that predicted their proliferative effects, without determining GREB1 levels (Fig 3E and F, lanes 5 and 14–16).	RESULTS
+9	19	cyclofenil	chemical	3,4‐DTP, cyclofenil, 3,4‐DTPD, and imidazopyridine analogs had NCOA1/3 recruitment profiles that predicted their proliferative effects, without determining GREB1 levels (Fig 3E and F, lanes 5 and 14–16).	RESULTS
+21	29	3,4‐DTPD	chemical	3,4‐DTP, cyclofenil, 3,4‐DTPD, and imidazopyridine analogs had NCOA1/3 recruitment profiles that predicted their proliferative effects, without determining GREB1 levels (Fig 3E and F, lanes 5 and 14–16).	RESULTS
+35	50	imidazopyridine	chemical	3,4‐DTP, cyclofenil, 3,4‐DTPD, and imidazopyridine analogs had NCOA1/3 recruitment profiles that predicted their proliferative effects, without determining GREB1 levels (Fig 3E and F, lanes 5 and 14–16).	RESULTS
+63	70	NCOA1/3	protein	3,4‐DTP, cyclofenil, 3,4‐DTPD, and imidazopyridine analogs had NCOA1/3 recruitment profiles that predicted their proliferative effects, without determining GREB1 levels (Fig 3E and F, lanes 5 and 14–16).	RESULTS
+156	161	GREB1	protein	3,4‐DTP, cyclofenil, 3,4‐DTPD, and imidazopyridine analogs had NCOA1/3 recruitment profiles that predicted their proliferative effects, without determining GREB1 levels (Fig 3E and F, lanes 5 and 14–16).	RESULTS
+11	19	S‐OBHS‐3	chemical	Similarly, S‐OBHS‐3, cyclofenil‐ASC, and OBHS‐ASC had positively correlated NCOA1/2/3 recruitment and GREB1 levels, but none of these activities determined their proliferative effects (Fig 3E and F lanes 11–12 and 18).	RESULTS
+21	35	cyclofenil‐ASC	chemical	Similarly, S‐OBHS‐3, cyclofenil‐ASC, and OBHS‐ASC had positively correlated NCOA1/2/3 recruitment and GREB1 levels, but none of these activities determined their proliferative effects (Fig 3E and F lanes 11–12 and 18).	RESULTS
+41	49	OBHS‐ASC	chemical	Similarly, S‐OBHS‐3, cyclofenil‐ASC, and OBHS‐ASC had positively correlated NCOA1/2/3 recruitment and GREB1 levels, but none of these activities determined their proliferative effects (Fig 3E and F lanes 11–12 and 18).	RESULTS
+76	85	NCOA1/2/3	protein	Similarly, S‐OBHS‐3, cyclofenil‐ASC, and OBHS‐ASC had positively correlated NCOA1/2/3 recruitment and GREB1 levels, but none of these activities determined their proliferative effects (Fig 3E and F lanes 11–12 and 18).	RESULTS
+102	107	GREB1	protein	Similarly, S‐OBHS‐3, cyclofenil‐ASC, and OBHS‐ASC had positively correlated NCOA1/2/3 recruitment and GREB1 levels, but none of these activities determined their proliferative effects (Fig 3E and F lanes 11–12 and 18).	RESULTS
+47	50	ERα	protein	For ligands that show cell‐specific signaling, ERα‐mediated recruitment of other coregulators and activation of other target genes likely determine their proliferative effects on MCF‐7 cells.	RESULTS
+0	5	NCOA3	protein	NCOA3 occupancy at GREB1 did not predict the proliferative response	RESULTS
+19	24	GREB1	protein	NCOA3 occupancy at GREB1 did not predict the proliferative response	RESULTS
+137	199	chromatin immunoprecipitation (ChIP)‐based quantitative assay,	experimental_method	We also questioned whether promoter occupancy by coactivators is statistically robust and reproducible for ligand class analysis using a chromatin immunoprecipitation (ChIP)‐based quantitative assay, and whether it has a better predictive power than the M2H assay.	RESULTS
+254	263	M2H assay	experimental_method	We also questioned whether promoter occupancy by coactivators is statistically robust and reproducible for ligand class analysis using a chromatin immunoprecipitation (ChIP)‐based quantitative assay, and whether it has a better predictive power than the M2H assay.	RESULTS
+0	3	ERα	protein	ERα and NCOA3 cycle on and off the GREB1 promoter (Nwachukwu et al, 2014).	RESULTS
+8	13	NCOA3	protein	ERα and NCOA3 cycle on and off the GREB1 promoter (Nwachukwu et al, 2014).	RESULTS
+35	40	GREB1	protein	ERα and NCOA3 cycle on and off the GREB1 promoter (Nwachukwu et al, 2014).	RESULTS
+32	49	time‐course study	experimental_method	Therefore, we first performed a time‐course study, and found that E2 and the WAY‐C analog, AAPII‐151‐4, induced recruitment of NCOA3 to the GREB1 promoter in a temporal cycle that peaked after 45 min in MCF‐7 cells (Fig 4A).	RESULTS
+66	68	E2	chemical	Therefore, we first performed a time‐course study, and found that E2 and the WAY‐C analog, AAPII‐151‐4, induced recruitment of NCOA3 to the GREB1 promoter in a temporal cycle that peaked after 45 min in MCF‐7 cells (Fig 4A).	RESULTS
+77	82	WAY‐C	chemical	Therefore, we first performed a time‐course study, and found that E2 and the WAY‐C analog, AAPII‐151‐4, induced recruitment of NCOA3 to the GREB1 promoter in a temporal cycle that peaked after 45 min in MCF‐7 cells (Fig 4A).	RESULTS
+91	102	AAPII‐151‐4	chemical	Therefore, we first performed a time‐course study, and found that E2 and the WAY‐C analog, AAPII‐151‐4, induced recruitment of NCOA3 to the GREB1 promoter in a temporal cycle that peaked after 45 min in MCF‐7 cells (Fig 4A).	RESULTS
+127	132	NCOA3	protein	Therefore, we first performed a time‐course study, and found that E2 and the WAY‐C analog, AAPII‐151‐4, induced recruitment of NCOA3 to the GREB1 promoter in a temporal cycle that peaked after 45 min in MCF‐7 cells (Fig 4A).	RESULTS
+140	145	GREB1	protein	Therefore, we first performed a time‐course study, and found that E2 and the WAY‐C analog, AAPII‐151‐4, induced recruitment of NCOA3 to the GREB1 promoter in a temporal cycle that peaked after 45 min in MCF‐7 cells (Fig 4A).	RESULTS
+26	31	WAY‐C	chemical	At this time point, other WAY‐C analogs also induced recruitment of NCOA3 at this site to varying degrees (Fig 4B).	RESULTS
+68	73	NCOA3	protein	At this time point, other WAY‐C analogs also induced recruitment of NCOA3 at this site to varying degrees (Fig 4B).	RESULTS
+4	6	Z’	evidence	The Z’ for this assay was 0.6, showing statistical robustness (see Materials and Methods).	RESULTS
+116	119	r 2	evidence	We prepared biological replicates with different cell passage numbers and separately prepared samples, which showed r 2 of 0.81, demonstrating high reproducibility (Fig 4C).	RESULTS
+1	6	NCOA3	protein	 NCOA3 occupancy at GREB1 is statistically robust but does not predict transcriptional activity	FIG
+20	25	GREB1	protein	 NCOA3 occupancy at GREB1 is statistically robust but does not predict transcriptional activity	FIG
+0	18	Kinetic ChIP assay	experimental_method	Kinetic ChIP assay examining recruitment of NCOA3 to the GREB1 gene in MCF‐7 cells stimulated with E2 or the indicated WAY‐C analog.	FIG
+44	49	NCOA3	protein	Kinetic ChIP assay examining recruitment of NCOA3 to the GREB1 gene in MCF‐7 cells stimulated with E2 or the indicated WAY‐C analog.	FIG
+57	62	GREB1	protein	Kinetic ChIP assay examining recruitment of NCOA3 to the GREB1 gene in MCF‐7 cells stimulated with E2 or the indicated WAY‐C analog.	FIG
+99	101	E2	chemical	Kinetic ChIP assay examining recruitment of NCOA3 to the GREB1 gene in MCF‐7 cells stimulated with E2 or the indicated WAY‐C analog.	FIG
+119	124	WAY‐C	chemical	Kinetic ChIP assay examining recruitment of NCOA3 to the GREB1 gene in MCF‐7 cells stimulated with E2 or the indicated WAY‐C analog.	FIG
+0	5	NCOA3	protein	NCOA3 occupancy at GREB1 was compared by ChIP assay 45 min after stimulation with vehicle, E2, or the WAY‐C analogs.	FIG
+19	24	GREB1	protein	NCOA3 occupancy at GREB1 was compared by ChIP assay 45 min after stimulation with vehicle, E2, or the WAY‐C analogs.	FIG
+41	51	ChIP assay	experimental_method	NCOA3 occupancy at GREB1 was compared by ChIP assay 45 min after stimulation with vehicle, E2, or the WAY‐C analogs.	FIG
+91	93	E2	chemical	NCOA3 occupancy at GREB1 was compared by ChIP assay 45 min after stimulation with vehicle, E2, or the WAY‐C analogs.	FIG
+102	107	WAY‐C	chemical	NCOA3 occupancy at GREB1 was compared by ChIP assay 45 min after stimulation with vehicle, E2, or the WAY‐C analogs.	FIG
+100	107	Z‐score	evidence	In panel (B), the average recruitment of two biological replicates are shown as mean + SEM, and the Z‐score is indicated.	FIG
+14	34	correlation analysis	experimental_method	In panel (C), correlation analysis was performed for two biological replicates.	FIG
+0	26	Linear regression analyses	experimental_method	Linear regression analyses comparing the ability of NCOA3 recruitment, measured by ChIP or M2H, to predict other agonist activities of WAY‐C analogs. *Significant positive correlation (F‐test for nonzero slope, P‐value).	FIG
+52	57	NCOA3	protein	Linear regression analyses comparing the ability of NCOA3 recruitment, measured by ChIP or M2H, to predict other agonist activities of WAY‐C analogs. *Significant positive correlation (F‐test for nonzero slope, P‐value).	FIG
+83	87	ChIP	experimental_method	Linear regression analyses comparing the ability of NCOA3 recruitment, measured by ChIP or M2H, to predict other agonist activities of WAY‐C analogs. *Significant positive correlation (F‐test for nonzero slope, P‐value).	FIG
+91	94	M2H	experimental_method	Linear regression analyses comparing the ability of NCOA3 recruitment, measured by ChIP or M2H, to predict other agonist activities of WAY‐C analogs. *Significant positive correlation (F‐test for nonzero slope, P‐value).	FIG
+135	140	WAY‐C	chemical	Linear regression analyses comparing the ability of NCOA3 recruitment, measured by ChIP or M2H, to predict other agonist activities of WAY‐C analogs. *Significant positive correlation (F‐test for nonzero slope, P‐value).	FIG
+185	191	F‐test	experimental_method	Linear regression analyses comparing the ability of NCOA3 recruitment, measured by ChIP or M2H, to predict other agonist activities of WAY‐C analogs. *Significant positive correlation (F‐test for nonzero slope, P‐value).	FIG
+211	218	P‐value	evidence	Linear regression analyses comparing the ability of NCOA3 recruitment, measured by ChIP or M2H, to predict other agonist activities of WAY‐C analogs. *Significant positive correlation (F‐test for nonzero slope, P‐value).	FIG
+4	13	M2H assay	experimental_method	The M2H assay for NCOA3 recruitment broadly correlated with the other assays, and was predictive for GREB1 expression and cell proliferation (Fig 3E).	RESULTS
+18	23	NCOA3	protein	The M2H assay for NCOA3 recruitment broadly correlated with the other assays, and was predictive for GREB1 expression and cell proliferation (Fig 3E).	RESULTS
+101	106	GREB1	protein	The M2H assay for NCOA3 recruitment broadly correlated with the other assays, and was predictive for GREB1 expression and cell proliferation (Fig 3E).	RESULTS
+13	24	ChIP assays	experimental_method	However, the ChIP assays for WAY‐C‐induced recruitment of NCOA3 to the GREB1 promoter did not correlate with any of the other WAY‐C activity profiles (Fig 4D), although the positive correlation between ChIP assays and NCOA3 recruitment via M2H assay showed a trend toward significance with r 2 = 0.36 and P = 0.09 (F‐test for nonzero slope).	RESULTS
+29	34	WAY‐C	chemical	However, the ChIP assays for WAY‐C‐induced recruitment of NCOA3 to the GREB1 promoter did not correlate with any of the other WAY‐C activity profiles (Fig 4D), although the positive correlation between ChIP assays and NCOA3 recruitment via M2H assay showed a trend toward significance with r 2 = 0.36 and P = 0.09 (F‐test for nonzero slope).	RESULTS
+58	63	NCOA3	protein	However, the ChIP assays for WAY‐C‐induced recruitment of NCOA3 to the GREB1 promoter did not correlate with any of the other WAY‐C activity profiles (Fig 4D), although the positive correlation between ChIP assays and NCOA3 recruitment via M2H assay showed a trend toward significance with r 2 = 0.36 and P = 0.09 (F‐test for nonzero slope).	RESULTS
+71	76	GREB1	protein	However, the ChIP assays for WAY‐C‐induced recruitment of NCOA3 to the GREB1 promoter did not correlate with any of the other WAY‐C activity profiles (Fig 4D), although the positive correlation between ChIP assays and NCOA3 recruitment via M2H assay showed a trend toward significance with r 2 = 0.36 and P = 0.09 (F‐test for nonzero slope).	RESULTS
+126	131	WAY‐C	chemical	However, the ChIP assays for WAY‐C‐induced recruitment of NCOA3 to the GREB1 promoter did not correlate with any of the other WAY‐C activity profiles (Fig 4D), although the positive correlation between ChIP assays and NCOA3 recruitment via M2H assay showed a trend toward significance with r 2 = 0.36 and P = 0.09 (F‐test for nonzero slope).	RESULTS
+202	213	ChIP assays	experimental_method	However, the ChIP assays for WAY‐C‐induced recruitment of NCOA3 to the GREB1 promoter did not correlate with any of the other WAY‐C activity profiles (Fig 4D), although the positive correlation between ChIP assays and NCOA3 recruitment via M2H assay showed a trend toward significance with r 2 = 0.36 and P = 0.09 (F‐test for nonzero slope).	RESULTS
+218	223	NCOA3	protein	However, the ChIP assays for WAY‐C‐induced recruitment of NCOA3 to the GREB1 promoter did not correlate with any of the other WAY‐C activity profiles (Fig 4D), although the positive correlation between ChIP assays and NCOA3 recruitment via M2H assay showed a trend toward significance with r 2 = 0.36 and P = 0.09 (F‐test for nonzero slope).	RESULTS
+240	249	M2H assay	experimental_method	However, the ChIP assays for WAY‐C‐induced recruitment of NCOA3 to the GREB1 promoter did not correlate with any of the other WAY‐C activity profiles (Fig 4D), although the positive correlation between ChIP assays and NCOA3 recruitment via M2H assay showed a trend toward significance with r 2 = 0.36 and P = 0.09 (F‐test for nonzero slope).	RESULTS
+290	293	r 2	evidence	However, the ChIP assays for WAY‐C‐induced recruitment of NCOA3 to the GREB1 promoter did not correlate with any of the other WAY‐C activity profiles (Fig 4D), although the positive correlation between ChIP assays and NCOA3 recruitment via M2H assay showed a trend toward significance with r 2 = 0.36 and P = 0.09 (F‐test for nonzero slope).	RESULTS
+305	306	P	evidence	However, the ChIP assays for WAY‐C‐induced recruitment of NCOA3 to the GREB1 promoter did not correlate with any of the other WAY‐C activity profiles (Fig 4D), although the positive correlation between ChIP assays and NCOA3 recruitment via M2H assay showed a trend toward significance with r 2 = 0.36 and P = 0.09 (F‐test for nonzero slope).	RESULTS
+315	321	F‐test	experimental_method	However, the ChIP assays for WAY‐C‐induced recruitment of NCOA3 to the GREB1 promoter did not correlate with any of the other WAY‐C activity profiles (Fig 4D), although the positive correlation between ChIP assays and NCOA3 recruitment via M2H assay showed a trend toward significance with r 2 = 0.36 and P = 0.09 (F‐test for nonzero slope).	RESULTS
+21	46	coactivator‐binding assay	experimental_method	Thus, the simplified coactivator‐binding assay showed much greater predictive power than the ChIP assay for ligand‐specific effects on GREB1 expression and cell proliferation.	RESULTS
+93	103	ChIP assay	experimental_method	Thus, the simplified coactivator‐binding assay showed much greater predictive power than the ChIP assay for ligand‐specific effects on GREB1 expression and cell proliferation.	RESULTS
+135	140	GREB1	protein	Thus, the simplified coactivator‐binding assay showed much greater predictive power than the ChIP assay for ligand‐specific effects on GREB1 expression and cell proliferation.	RESULTS
+0	3	ERβ	protein	ERβ activity is not an independent predictor of cell‐specific activity	RESULTS
+110	113	ERβ	protein	One difference between MCF‐7 breast cancer cells and Ishikawa endometrial cancer cells is the contribution of ERβ to estrogenic response, as Ishikawa cells may express ERβ (Bhat & Pezzuto, 2001).	RESULTS
+168	171	ERβ	protein	One difference between MCF‐7 breast cancer cells and Ishikawa endometrial cancer cells is the contribution of ERβ to estrogenic response, as Ishikawa cells may express ERβ (Bhat & Pezzuto, 2001).	RESULTS
+5	18	overexpressed	experimental_method	When overexpressed in MCF‐7 cells, ERβ alters E2‐induced expression of only a subset of ERα‐target genes (Wu et al, 2011), raising the possibility that ligand‐induced ERβ activity may contribute to E‐Luc activities, and thus underlie the lack of correlation between the E‐Luc and L‐Luc ERα‐WT activities or GREB1 levels induced by cell‐specific modulators in cluster 2 and cluster 3 (Fig 3C).	RESULTS
+35	38	ERβ	protein	When overexpressed in MCF‐7 cells, ERβ alters E2‐induced expression of only a subset of ERα‐target genes (Wu et al, 2011), raising the possibility that ligand‐induced ERβ activity may contribute to E‐Luc activities, and thus underlie the lack of correlation between the E‐Luc and L‐Luc ERα‐WT activities or GREB1 levels induced by cell‐specific modulators in cluster 2 and cluster 3 (Fig 3C).	RESULTS
+46	48	E2	chemical	When overexpressed in MCF‐7 cells, ERβ alters E2‐induced expression of only a subset of ERα‐target genes (Wu et al, 2011), raising the possibility that ligand‐induced ERβ activity may contribute to E‐Luc activities, and thus underlie the lack of correlation between the E‐Luc and L‐Luc ERα‐WT activities or GREB1 levels induced by cell‐specific modulators in cluster 2 and cluster 3 (Fig 3C).	RESULTS
+88	91	ERα	protein	When overexpressed in MCF‐7 cells, ERβ alters E2‐induced expression of only a subset of ERα‐target genes (Wu et al, 2011), raising the possibility that ligand‐induced ERβ activity may contribute to E‐Luc activities, and thus underlie the lack of correlation between the E‐Luc and L‐Luc ERα‐WT activities or GREB1 levels induced by cell‐specific modulators in cluster 2 and cluster 3 (Fig 3C).	RESULTS
+167	170	ERβ	protein	When overexpressed in MCF‐7 cells, ERβ alters E2‐induced expression of only a subset of ERα‐target genes (Wu et al, 2011), raising the possibility that ligand‐induced ERβ activity may contribute to E‐Luc activities, and thus underlie the lack of correlation between the E‐Luc and L‐Luc ERα‐WT activities or GREB1 levels induced by cell‐specific modulators in cluster 2 and cluster 3 (Fig 3C).	RESULTS
+198	203	E‐Luc	experimental_method	When overexpressed in MCF‐7 cells, ERβ alters E2‐induced expression of only a subset of ERα‐target genes (Wu et al, 2011), raising the possibility that ligand‐induced ERβ activity may contribute to E‐Luc activities, and thus underlie the lack of correlation between the E‐Luc and L‐Luc ERα‐WT activities or GREB1 levels induced by cell‐specific modulators in cluster 2 and cluster 3 (Fig 3C).	RESULTS
+270	275	E‐Luc	experimental_method	When overexpressed in MCF‐7 cells, ERβ alters E2‐induced expression of only a subset of ERα‐target genes (Wu et al, 2011), raising the possibility that ligand‐induced ERβ activity may contribute to E‐Luc activities, and thus underlie the lack of correlation between the E‐Luc and L‐Luc ERα‐WT activities or GREB1 levels induced by cell‐specific modulators in cluster 2 and cluster 3 (Fig 3C).	RESULTS
+280	285	L‐Luc	experimental_method	When overexpressed in MCF‐7 cells, ERβ alters E2‐induced expression of only a subset of ERα‐target genes (Wu et al, 2011), raising the possibility that ligand‐induced ERβ activity may contribute to E‐Luc activities, and thus underlie the lack of correlation between the E‐Luc and L‐Luc ERα‐WT activities or GREB1 levels induced by cell‐specific modulators in cluster 2 and cluster 3 (Fig 3C).	RESULTS
+286	289	ERα	protein	When overexpressed in MCF‐7 cells, ERβ alters E2‐induced expression of only a subset of ERα‐target genes (Wu et al, 2011), raising the possibility that ligand‐induced ERβ activity may contribute to E‐Luc activities, and thus underlie the lack of correlation between the E‐Luc and L‐Luc ERα‐WT activities or GREB1 levels induced by cell‐specific modulators in cluster 2 and cluster 3 (Fig 3C).	RESULTS
+290	292	WT	protein_state	When overexpressed in MCF‐7 cells, ERβ alters E2‐induced expression of only a subset of ERα‐target genes (Wu et al, 2011), raising the possibility that ligand‐induced ERβ activity may contribute to E‐Luc activities, and thus underlie the lack of correlation between the E‐Luc and L‐Luc ERα‐WT activities or GREB1 levels induced by cell‐specific modulators in cluster 2 and cluster 3 (Fig 3C).	RESULTS
+307	312	GREB1	protein	When overexpressed in MCF‐7 cells, ERβ alters E2‐induced expression of only a subset of ERα‐target genes (Wu et al, 2011), raising the possibility that ligand‐induced ERβ activity may contribute to E‐Luc activities, and thus underlie the lack of correlation between the E‐Luc and L‐Luc ERα‐WT activities or GREB1 levels induced by cell‐specific modulators in cluster 2 and cluster 3 (Fig 3C).	RESULTS
+37	42	L‐Luc	experimental_method	To test this idea, we determined the L‐Luc ERβ activity profiles of the ligands (Fig EV1).	RESULTS
+59	63	OBHS	chemical	All direct modulator and two indirect modulator scaffolds (OBHS and S‐OBHS‐3) lacked ERβ agonist activity.	RESULTS
+68	76	S‐OBHS‐3	chemical	All direct modulator and two indirect modulator scaffolds (OBHS and S‐OBHS‐3) lacked ERβ agonist activity.	RESULTS
+20	25	L‐Luc	experimental_method	For most scaffolds, L‐Luc ERβ and E‐Luc activities were not correlated, except for 2,5‐DTP and cyclofenil analogs, which showed moderate but significant correlations (Fig EV4A).	RESULTS
+34	39	E‐Luc	experimental_method	For most scaffolds, L‐Luc ERβ and E‐Luc activities were not correlated, except for 2,5‐DTP and cyclofenil analogs, which showed moderate but significant correlations (Fig EV4A).	RESULTS
+83	90	2,5‐DTP	chemical	For most scaffolds, L‐Luc ERβ and E‐Luc activities were not correlated, except for 2,5‐DTP and cyclofenil analogs, which showed moderate but significant correlations (Fig EV4A).	RESULTS
+95	105	cyclofenil	chemical	For most scaffolds, L‐Luc ERβ and E‐Luc activities were not correlated, except for 2,5‐DTP and cyclofenil analogs, which showed moderate but significant correlations (Fig EV4A).	RESULTS
+18	23	E‐Luc	experimental_method	Nevertheless, the E‐Luc activities of both 2,5‐DTP and cyclofenil analogs were better predicted by their L‐Luc ERα‐WT than L‐Luc ERβ activities (Fig EV4A and B).	RESULTS
+43	50	2,5‐DTP	chemical	Nevertheless, the E‐Luc activities of both 2,5‐DTP and cyclofenil analogs were better predicted by their L‐Luc ERα‐WT than L‐Luc ERβ activities (Fig EV4A and B).	RESULTS
+55	65	cyclofenil	chemical	Nevertheless, the E‐Luc activities of both 2,5‐DTP and cyclofenil analogs were better predicted by their L‐Luc ERα‐WT than L‐Luc ERβ activities (Fig EV4A and B).	RESULTS
+105	110	L‐Luc	experimental_method	Nevertheless, the E‐Luc activities of both 2,5‐DTP and cyclofenil analogs were better predicted by their L‐Luc ERα‐WT than L‐Luc ERβ activities (Fig EV4A and B).	RESULTS
+111	114	ERα	protein	Nevertheless, the E‐Luc activities of both 2,5‐DTP and cyclofenil analogs were better predicted by their L‐Luc ERα‐WT than L‐Luc ERβ activities (Fig EV4A and B).	RESULTS
+115	117	WT	protein_state	Nevertheless, the E‐Luc activities of both 2,5‐DTP and cyclofenil analogs were better predicted by their L‐Luc ERα‐WT than L‐Luc ERβ activities (Fig EV4A and B).	RESULTS
+123	128	L‐Luc	experimental_method	Nevertheless, the E‐Luc activities of both 2,5‐DTP and cyclofenil analogs were better predicted by their L‐Luc ERα‐WT than L‐Luc ERβ activities (Fig EV4A and B).	RESULTS
+129	132	ERβ	protein	Nevertheless, the E‐Luc activities of both 2,5‐DTP and cyclofenil analogs were better predicted by their L‐Luc ERα‐WT than L‐Luc ERβ activities (Fig EV4A and B).	RESULTS
+0	3	ERβ	protein	ERβ activity is not an independent predictor of E‐Luc activity	FIG
+48	53	E‐Luc	experimental_method	ERβ activity is not an independent predictor of E‐Luc activity	FIG
+0	3	ERβ	protein	ERβ activity in HepG2 cells rarely correlates with E‐Luc activity.	FIG
+51	56	E‐Luc	experimental_method	ERβ activity in HepG2 cells rarely correlates with E‐Luc activity.	FIG
+0	3	ERα	protein	ERα activity of 2,5‐DTP and cyclofenil analogs correlates with E‐Luc activity.	FIG
+16	23	2,5‐DTP	chemical	ERα activity of 2,5‐DTP and cyclofenil analogs correlates with E‐Luc activity.	FIG
+28	38	cyclofenil	chemical	ERα activity of 2,5‐DTP and cyclofenil analogs correlates with E‐Luc activity.	FIG
+63	68	E‐Luc	experimental_method	ERα activity of 2,5‐DTP and cyclofenil analogs correlates with E‐Luc activity.	FIG
+31	39	P values	evidence	 Data information: The r 2 and P values for the indicated correlations are shown in both panels. *Significant positive correlation (F‐test for nonzero slope, P‐value)	FIG
+132	138	F‐test	experimental_method	 Data information: The r 2 and P values for the indicated correlations are shown in both panels. *Significant positive correlation (F‐test for nonzero slope, P‐value)	FIG
+158	165	P‐value	evidence	 Data information: The r 2 and P values for the indicated correlations are shown in both panels. *Significant positive correlation (F‐test for nonzero slope, P‐value)	FIG
+24	39	crystallization	experimental_method	To overcome barriers to crystallization of ERα LBD complexes, we developed a conformation‐trapping X‐ray crystallography approach using the ERα‐Y537S mutation (Nettles et al, 2008; Bruning et al, 2010; Srinivasan et al, 2013).	RESULTS
+43	46	ERα	protein	To overcome barriers to crystallization of ERα LBD complexes, we developed a conformation‐trapping X‐ray crystallography approach using the ERα‐Y537S mutation (Nettles et al, 2008; Bruning et al, 2010; Srinivasan et al, 2013).	RESULTS
+47	50	LBD	structure_element	To overcome barriers to crystallization of ERα LBD complexes, we developed a conformation‐trapping X‐ray crystallography approach using the ERα‐Y537S mutation (Nettles et al, 2008; Bruning et al, 2010; Srinivasan et al, 2013).	RESULTS
+77	120	conformation‐trapping X‐ray crystallography	experimental_method	To overcome barriers to crystallization of ERα LBD complexes, we developed a conformation‐trapping X‐ray crystallography approach using the ERα‐Y537S mutation (Nettles et al, 2008; Bruning et al, 2010; Srinivasan et al, 2013).	RESULTS
+140	149	ERα‐Y537S	mutant	To overcome barriers to crystallization of ERα LBD complexes, we developed a conformation‐trapping X‐ray crystallography approach using the ERα‐Y537S mutation (Nettles et al, 2008; Bruning et al, 2010; Srinivasan et al, 2013).	RESULTS
+38	44	solved	experimental_method	To further validate this approach, we solved the structure of the ERα‐Y537S LBD in complex with diethylstilbestrol (DES), which bound identically in the wild‐type and ERα‐Y537S LBDs, demonstrating again that this surface mutation stabilizes h12 dynamics to facilitate crystallization without changing ligand binding (Appendix Fig S1A and B) (Nettles et al, 2008; Bruning et al, 2010; Delfosse et al, 2012).	RESULTS
+49	58	structure	evidence	To further validate this approach, we solved the structure of the ERα‐Y537S LBD in complex with diethylstilbestrol (DES), which bound identically in the wild‐type and ERα‐Y537S LBDs, demonstrating again that this surface mutation stabilizes h12 dynamics to facilitate crystallization without changing ligand binding (Appendix Fig S1A and B) (Nettles et al, 2008; Bruning et al, 2010; Delfosse et al, 2012).	RESULTS
+66	75	ERα‐Y537S	mutant	To further validate this approach, we solved the structure of the ERα‐Y537S LBD in complex with diethylstilbestrol (DES), which bound identically in the wild‐type and ERα‐Y537S LBDs, demonstrating again that this surface mutation stabilizes h12 dynamics to facilitate crystallization without changing ligand binding (Appendix Fig S1A and B) (Nettles et al, 2008; Bruning et al, 2010; Delfosse et al, 2012).	RESULTS
+76	79	LBD	structure_element	To further validate this approach, we solved the structure of the ERα‐Y537S LBD in complex with diethylstilbestrol (DES), which bound identically in the wild‐type and ERα‐Y537S LBDs, demonstrating again that this surface mutation stabilizes h12 dynamics to facilitate crystallization without changing ligand binding (Appendix Fig S1A and B) (Nettles et al, 2008; Bruning et al, 2010; Delfosse et al, 2012).	RESULTS
+80	95	in complex with	protein_state	To further validate this approach, we solved the structure of the ERα‐Y537S LBD in complex with diethylstilbestrol (DES), which bound identically in the wild‐type and ERα‐Y537S LBDs, demonstrating again that this surface mutation stabilizes h12 dynamics to facilitate crystallization without changing ligand binding (Appendix Fig S1A and B) (Nettles et al, 2008; Bruning et al, 2010; Delfosse et al, 2012).	RESULTS
+96	114	diethylstilbestrol	chemical	To further validate this approach, we solved the structure of the ERα‐Y537S LBD in complex with diethylstilbestrol (DES), which bound identically in the wild‐type and ERα‐Y537S LBDs, demonstrating again that this surface mutation stabilizes h12 dynamics to facilitate crystallization without changing ligand binding (Appendix Fig S1A and B) (Nettles et al, 2008; Bruning et al, 2010; Delfosse et al, 2012).	RESULTS
+116	119	DES	chemical	To further validate this approach, we solved the structure of the ERα‐Y537S LBD in complex with diethylstilbestrol (DES), which bound identically in the wild‐type and ERα‐Y537S LBDs, demonstrating again that this surface mutation stabilizes h12 dynamics to facilitate crystallization without changing ligand binding (Appendix Fig S1A and B) (Nettles et al, 2008; Bruning et al, 2010; Delfosse et al, 2012).	RESULTS
+153	162	wild‐type	protein_state	To further validate this approach, we solved the structure of the ERα‐Y537S LBD in complex with diethylstilbestrol (DES), which bound identically in the wild‐type and ERα‐Y537S LBDs, demonstrating again that this surface mutation stabilizes h12 dynamics to facilitate crystallization without changing ligand binding (Appendix Fig S1A and B) (Nettles et al, 2008; Bruning et al, 2010; Delfosse et al, 2012).	RESULTS
+167	176	ERα‐Y537S	mutant	To further validate this approach, we solved the structure of the ERα‐Y537S LBD in complex with diethylstilbestrol (DES), which bound identically in the wild‐type and ERα‐Y537S LBDs, demonstrating again that this surface mutation stabilizes h12 dynamics to facilitate crystallization without changing ligand binding (Appendix Fig S1A and B) (Nettles et al, 2008; Bruning et al, 2010; Delfosse et al, 2012).	RESULTS
+177	181	LBDs	structure_element	To further validate this approach, we solved the structure of the ERα‐Y537S LBD in complex with diethylstilbestrol (DES), which bound identically in the wild‐type and ERα‐Y537S LBDs, demonstrating again that this surface mutation stabilizes h12 dynamics to facilitate crystallization without changing ligand binding (Appendix Fig S1A and B) (Nettles et al, 2008; Bruning et al, 2010; Delfosse et al, 2012).	RESULTS
+241	244	h12	structure_element	To further validate this approach, we solved the structure of the ERα‐Y537S LBD in complex with diethylstilbestrol (DES), which bound identically in the wild‐type and ERα‐Y537S LBDs, demonstrating again that this surface mutation stabilizes h12 dynamics to facilitate crystallization without changing ligand binding (Appendix Fig S1A and B) (Nettles et al, 2008; Bruning et al, 2010; Delfosse et al, 2012).	RESULTS
+24	30	solved	experimental_method	Using this approach, we solved 76 ERα LBD structures in the active conformation and bound to ligands studied here (Appendix Fig S1C).	RESULTS
+34	37	ERα	protein	Using this approach, we solved 76 ERα LBD structures in the active conformation and bound to ligands studied here (Appendix Fig S1C).	RESULTS
+38	41	LBD	structure_element	Using this approach, we solved 76 ERα LBD structures in the active conformation and bound to ligands studied here (Appendix Fig S1C).	RESULTS
+42	52	structures	evidence	Using this approach, we solved 76 ERα LBD structures in the active conformation and bound to ligands studied here (Appendix Fig S1C).	RESULTS
+60	79	active conformation	protein_state	Using this approach, we solved 76 ERα LBD structures in the active conformation and bound to ligands studied here (Appendix Fig S1C).	RESULTS
+84	100	bound to ligands	protein_state	Using this approach, we solved 76 ERα LBD structures in the active conformation and bound to ligands studied here (Appendix Fig S1C).	RESULTS
+16	26	structures	evidence	Eleven of these structures have been published, while 65 are new, including the DES‐bound ERα‐Y537S LBD.	RESULTS
+80	89	DES‐bound	protein_state	Eleven of these structures have been published, while 65 are new, including the DES‐bound ERα‐Y537S LBD.	RESULTS
+90	99	ERα‐Y537S	mutant	Eleven of these structures have been published, while 65 are new, including the DES‐bound ERα‐Y537S LBD.	RESULTS
+100	103	LBD	structure_element	Eleven of these structures have been published, while 65 are new, including the DES‐bound ERα‐Y537S LBD.	RESULTS
+27	37	structures	evidence	We present 57 of these new structures here (Dataset EV2), while the remaining eight new structures bound to OBHS‐N analogs will be published elsewhere (S. Srinivasan et al, in preparation).	RESULTS
+88	98	structures	evidence	We present 57 of these new structures here (Dataset EV2), while the remaining eight new structures bound to OBHS‐N analogs will be published elsewhere (S. Srinivasan et al, in preparation).	RESULTS
+99	107	bound to	protein_state	We present 57 of these new structures here (Dataset EV2), while the remaining eight new structures bound to OBHS‐N analogs will be published elsewhere (S. Srinivasan et al, in preparation).	RESULTS
+108	114	OBHS‐N	chemical	We present 57 of these new structures here (Dataset EV2), while the remaining eight new structures bound to OBHS‐N analogs will be published elsewhere (S. Srinivasan et al, in preparation).	RESULTS
+31	41	structures	evidence	Examining many closely related structures allows us to visualize subtle structural differences, in effect using X‐ray crystallography as a systems biology tool.	RESULTS
+112	133	X‐ray crystallography	experimental_method	Examining many closely related structures allows us to visualize subtle structural differences, in effect using X‐ray crystallography as a systems biology tool.	RESULTS
+165	168	h12	structure_element	The indirect modulator scaffolds in cluster 1 did not show cell‐specific signaling (Fig 3C), but shared common structural perturbations that we designed to modulate h12 dynamics.	RESULTS
+22	26	OBHS	chemical	Based on our original OBHS structure, the OBHS, OBHS‐N, and triaryl‐ethylene compounds were modified with h11‐directed pendant groups (Zheng et al, 2012; Zhu et al, 2012; Liao et al, 2014).	RESULTS
+27	36	structure	evidence	Based on our original OBHS structure, the OBHS, OBHS‐N, and triaryl‐ethylene compounds were modified with h11‐directed pendant groups (Zheng et al, 2012; Zhu et al, 2012; Liao et al, 2014).	RESULTS
+42	46	OBHS	chemical	Based on our original OBHS structure, the OBHS, OBHS‐N, and triaryl‐ethylene compounds were modified with h11‐directed pendant groups (Zheng et al, 2012; Zhu et al, 2012; Liao et al, 2014).	RESULTS
+48	54	OBHS‐N	chemical	Based on our original OBHS structure, the OBHS, OBHS‐N, and triaryl‐ethylene compounds were modified with h11‐directed pendant groups (Zheng et al, 2012; Zhu et al, 2012; Liao et al, 2014).	RESULTS
+60	76	triaryl‐ethylene	chemical	Based on our original OBHS structure, the OBHS, OBHS‐N, and triaryl‐ethylene compounds were modified with h11‐directed pendant groups (Zheng et al, 2012; Zhu et al, 2012; Liao et al, 2014).	RESULTS
+106	109	h11	structure_element	Based on our original OBHS structure, the OBHS, OBHS‐N, and triaryl‐ethylene compounds were modified with h11‐directed pendant groups (Zheng et al, 2012; Zhu et al, 2012; Liao et al, 2014).	RESULTS
+0	11	Superposing	experimental_method	Superposing the LBDs based on the class of bound ligands provides an ensemble view of the structural variance and clarifies what part of the ligand‐binding pocket is differentially perturbed or targeted.	RESULTS
+16	20	LBDs	structure_element	Superposing the LBDs based on the class of bound ligands provides an ensemble view of the structural variance and clarifies what part of the ligand‐binding pocket is differentially perturbed or targeted.	RESULTS
+141	162	ligand‐binding pocket	site	Superposing the LBDs based on the class of bound ligands provides an ensemble view of the structural variance and clarifies what part of the ligand‐binding pocket is differentially perturbed or targeted.	RESULTS
+7	17	structures	evidence	The 24 structures containing OBHS, OBHS‐N, or triaryl‐ethylene analogs showed structural diversity in the same part of the scaffolds (Figs 5A and EV5A), and the same region of the LBD—the C‐terminal end of h11 (Figs 5B and C, and EV5B), which in turn nudges h12 (Fig 5C and D).	RESULTS
+29	33	OBHS	chemical	The 24 structures containing OBHS, OBHS‐N, or triaryl‐ethylene analogs showed structural diversity in the same part of the scaffolds (Figs 5A and EV5A), and the same region of the LBD—the C‐terminal end of h11 (Figs 5B and C, and EV5B), which in turn nudges h12 (Fig 5C and D).	RESULTS
+35	41	OBHS‐N	chemical	The 24 structures containing OBHS, OBHS‐N, or triaryl‐ethylene analogs showed structural diversity in the same part of the scaffolds (Figs 5A and EV5A), and the same region of the LBD—the C‐terminal end of h11 (Figs 5B and C, and EV5B), which in turn nudges h12 (Fig 5C and D).	RESULTS
+46	62	triaryl‐ethylene	chemical	The 24 structures containing OBHS, OBHS‐N, or triaryl‐ethylene analogs showed structural diversity in the same part of the scaffolds (Figs 5A and EV5A), and the same region of the LBD—the C‐terminal end of h11 (Figs 5B and C, and EV5B), which in turn nudges h12 (Fig 5C and D).	RESULTS
+180	183	LBD	structure_element	The 24 structures containing OBHS, OBHS‐N, or triaryl‐ethylene analogs showed structural diversity in the same part of the scaffolds (Figs 5A and EV5A), and the same region of the LBD—the C‐terminal end of h11 (Figs 5B and C, and EV5B), which in turn nudges h12 (Fig 5C and D).	RESULTS
+206	209	h11	structure_element	The 24 structures containing OBHS, OBHS‐N, or triaryl‐ethylene analogs showed structural diversity in the same part of the scaffolds (Figs 5A and EV5A), and the same region of the LBD—the C‐terminal end of h11 (Figs 5B and C, and EV5B), which in turn nudges h12 (Fig 5C and D).	RESULTS
+258	261	h12	structure_element	The 24 structures containing OBHS, OBHS‐N, or triaryl‐ethylene analogs showed structural diversity in the same part of the scaffolds (Figs 5A and EV5A), and the same region of the LBD—the C‐terminal end of h11 (Figs 5B and C, and EV5B), which in turn nudges h12 (Fig 5C and D).	RESULTS
+21	27	OBHS‐N	chemical	We observed that the OBHS‐N analogs displaced h11 along a vector away from Leu354 in a region of h3 that is unaffected by the ligands, and toward the dimer interface.	RESULTS
+46	49	h11	structure_element	We observed that the OBHS‐N analogs displaced h11 along a vector away from Leu354 in a region of h3 that is unaffected by the ligands, and toward the dimer interface.	RESULTS
+75	81	Leu354	residue_name_number	We observed that the OBHS‐N analogs displaced h11 along a vector away from Leu354 in a region of h3 that is unaffected by the ligands, and toward the dimer interface.	RESULTS
+97	99	h3	structure_element	We observed that the OBHS‐N analogs displaced h11 along a vector away from Leu354 in a region of h3 that is unaffected by the ligands, and toward the dimer interface.	RESULTS
+150	165	dimer interface	site	We observed that the OBHS‐N analogs displaced h11 along a vector away from Leu354 in a region of h3 that is unaffected by the ligands, and toward the dimer interface.	RESULTS
+8	24	triaryl‐ethylene	chemical	For the triaryl‐ethylene analogs, the displacement of h11 was in a perpendicular direction, away from Ile424 in h8 and toward h12.	RESULTS
+54	57	h11	structure_element	For the triaryl‐ethylene analogs, the displacement of h11 was in a perpendicular direction, away from Ile424 in h8 and toward h12.	RESULTS
+102	108	Ile424	residue_name_number	For the triaryl‐ethylene analogs, the displacement of h11 was in a perpendicular direction, away from Ile424 in h8 and toward h12.	RESULTS
+112	114	h8	structure_element	For the triaryl‐ethylene analogs, the displacement of h11 was in a perpendicular direction, away from Ile424 in h8 and toward h12.	RESULTS
+126	129	h12	structure_element	For the triaryl‐ethylene analogs, the displacement of h11 was in a perpendicular direction, away from Ile424 in h8 and toward h12.	RESULTS
+29	51	inter‐atomic distances	evidence	Remarkably, these individual inter‐atomic distances showed a ligand class‐specific ability to significantly predict proliferative effects (Fig 5E and F), demonstrating the feasibility of developing a minimal set of activity predictors from crystal structures.	RESULTS
+240	258	crystal structures	evidence	Remarkably, these individual inter‐atomic distances showed a ligand class‐specific ability to significantly predict proliferative effects (Fig 5E and F), demonstrating the feasibility of developing a minimal set of activity predictors from crystal structures.	RESULTS
+0	24	Structure‐class analysis	experimental_method	Structure‐class analysis of triaryl‐ethylene analogs.	FIG
+28	44	triaryl‐ethylene	chemical	Structure‐class analysis of triaryl‐ethylene analogs.	FIG
+0	16	Triaryl‐ethylene	chemical	Triaryl‐ethylene analogs bound to the superposed crystal structures of the ERα LBD are shown.	FIG
+25	33	bound to	protein_state	Triaryl‐ethylene analogs bound to the superposed crystal structures of the ERα LBD are shown.	FIG
+38	48	superposed	experimental_method	Triaryl‐ethylene analogs bound to the superposed crystal structures of the ERα LBD are shown.	FIG
+49	67	crystal structures	evidence	Triaryl‐ethylene analogs bound to the superposed crystal structures of the ERα LBD are shown.	FIG
+75	78	ERα	protein	Triaryl‐ethylene analogs bound to the superposed crystal structures of the ERα LBD are shown.	FIG
+79	82	LBD	structure_element	Triaryl‐ethylene analogs bound to the superposed crystal structures of the ERα LBD are shown.	FIG
+70	73	h11	structure_element	Arrows indicate chemical variance in the orientation of the different h11‐directed ligand side groups (PDB 5DK9, 5DKB, 5DKE, 5DKG, 5DKS, 5DL4, 5DLR, 5DMC, 5DMF and 5DP0).	FIG
+0	16	Triaryl‐ethylene	chemical	Triaryl‐ethylene analogs induce variance of ERα conformations at the C‐terminal region of h11.	FIG
+44	47	ERα	protein	Triaryl‐ethylene analogs induce variance of ERα conformations at the C‐terminal region of h11.	FIG
+90	93	h11	structure_element	Triaryl‐ethylene analogs induce variance of ERα conformations at the C‐terminal region of h11.	FIG
+20	37	crystal structure	evidence	Panel (B) shows the crystal structure of a triaryl‐ethylene analog‐bound ERα LBD (PDB 5DLR).	FIG
+43	59	triaryl‐ethylene	chemical	Panel (B) shows the crystal structure of a triaryl‐ethylene analog‐bound ERα LBD (PDB 5DLR).	FIG
+73	76	ERα	protein	Panel (B) shows the crystal structure of a triaryl‐ethylene analog‐bound ERα LBD (PDB 5DLR).	FIG
+77	80	LBD	structure_element	Panel (B) shows the crystal structure of a triaryl‐ethylene analog‐bound ERα LBD (PDB 5DLR).	FIG
+4	21	h11–h12 interface	site	The h11–h12 interface (circled) includes the C‐terminal part of h11.	FIG
+64	67	h11	structure_element	The h11–h12 interface (circled) includes the C‐terminal part of h11.	FIG
+52	68	triaryl‐ethylene	chemical	This region was expanded in panel (C), where the 10 triaryl‐ethylene analog‐bound ERα LBD structures (see Datasets EV1 and EV2) were superposed to show variations in the h11 C‐terminus (PDB 5DK9, 5DKB, 5DKE, 5DKG, 5DKS, 5DL4, 5DLR, 5DMC, 5DMF, and 5DP0).	FIG
+82	85	ERα	protein	This region was expanded in panel (C), where the 10 triaryl‐ethylene analog‐bound ERα LBD structures (see Datasets EV1 and EV2) were superposed to show variations in the h11 C‐terminus (PDB 5DK9, 5DKB, 5DKE, 5DKG, 5DKS, 5DL4, 5DLR, 5DMC, 5DMF, and 5DP0).	FIG
+86	89	LBD	structure_element	This region was expanded in panel (C), where the 10 triaryl‐ethylene analog‐bound ERα LBD structures (see Datasets EV1 and EV2) were superposed to show variations in the h11 C‐terminus (PDB 5DK9, 5DKB, 5DKE, 5DKG, 5DKS, 5DL4, 5DLR, 5DMC, 5DMF, and 5DP0).	FIG
+90	100	structures	evidence	This region was expanded in panel (C), where the 10 triaryl‐ethylene analog‐bound ERα LBD structures (see Datasets EV1 and EV2) were superposed to show variations in the h11 C‐terminus (PDB 5DK9, 5DKB, 5DKE, 5DKG, 5DKS, 5DL4, 5DLR, 5DMC, 5DMF, and 5DP0).	FIG
+133	143	superposed	experimental_method	This region was expanded in panel (C), where the 10 triaryl‐ethylene analog‐bound ERα LBD structures (see Datasets EV1 and EV2) were superposed to show variations in the h11 C‐terminus (PDB 5DK9, 5DKB, 5DKE, 5DKG, 5DKS, 5DL4, 5DLR, 5DMC, 5DMF, and 5DP0).	FIG
+170	173	h11	structure_element	This region was expanded in panel (C), where the 10 triaryl‐ethylene analog‐bound ERα LBD structures (see Datasets EV1 and EV2) were superposed to show variations in the h11 C‐terminus (PDB 5DK9, 5DKB, 5DKE, 5DKG, 5DKS, 5DL4, 5DLR, 5DMC, 5DMF, and 5DP0).	FIG
+0	3	ERα	protein	ERα LBDs in complex with diethylstilbestrol (DES) or a triaryl‐ethylene analog were superposed to show that the ligand‐induced difference in h11 conformation is transmitted to the C‐terminus of h12 (PDB 4ZN7, 5DMC).	FIG
+4	8	LBDs	structure_element	ERα LBDs in complex with diethylstilbestrol (DES) or a triaryl‐ethylene analog were superposed to show that the ligand‐induced difference in h11 conformation is transmitted to the C‐terminus of h12 (PDB 4ZN7, 5DMC).	FIG
+9	24	in complex with	protein_state	ERα LBDs in complex with diethylstilbestrol (DES) or a triaryl‐ethylene analog were superposed to show that the ligand‐induced difference in h11 conformation is transmitted to the C‐terminus of h12 (PDB 4ZN7, 5DMC).	FIG
+25	43	diethylstilbestrol	chemical	ERα LBDs in complex with diethylstilbestrol (DES) or a triaryl‐ethylene analog were superposed to show that the ligand‐induced difference in h11 conformation is transmitted to the C‐terminus of h12 (PDB 4ZN7, 5DMC).	FIG
+45	48	DES	chemical	ERα LBDs in complex with diethylstilbestrol (DES) or a triaryl‐ethylene analog were superposed to show that the ligand‐induced difference in h11 conformation is transmitted to the C‐terminus of h12 (PDB 4ZN7, 5DMC).	FIG
+55	71	triaryl‐ethylene	chemical	ERα LBDs in complex with diethylstilbestrol (DES) or a triaryl‐ethylene analog were superposed to show that the ligand‐induced difference in h11 conformation is transmitted to the C‐terminus of h12 (PDB 4ZN7, 5DMC).	FIG
+84	94	superposed	experimental_method	ERα LBDs in complex with diethylstilbestrol (DES) or a triaryl‐ethylene analog were superposed to show that the ligand‐induced difference in h11 conformation is transmitted to the C‐terminus of h12 (PDB 4ZN7, 5DMC).	FIG
+141	144	h11	structure_element	ERα LBDs in complex with diethylstilbestrol (DES) or a triaryl‐ethylene analog were superposed to show that the ligand‐induced difference in h11 conformation is transmitted to the C‐terminus of h12 (PDB 4ZN7, 5DMC).	FIG
+194	197	h12	structure_element	ERα LBDs in complex with diethylstilbestrol (DES) or a triaryl‐ethylene analog were superposed to show that the ligand‐induced difference in h11 conformation is transmitted to the C‐terminus of h12 (PDB 4ZN7, 5DMC).	FIG
+0	22	Inter‐atomic distances	evidence	Inter‐atomic distances predict the proliferative effects of specific ligand series.	FIG
+0	6	Ile424	residue_name_number	Ile424–His524 distance measured in the crystal structures correlates with the proliferative effect of triaryl‐ethylene analogs in MCF‐7 cells.	FIG
+7	13	His524	residue_name_number	Ile424–His524 distance measured in the crystal structures correlates with the proliferative effect of triaryl‐ethylene analogs in MCF‐7 cells.	FIG
+14	22	distance	evidence	Ile424–His524 distance measured in the crystal structures correlates with the proliferative effect of triaryl‐ethylene analogs in MCF‐7 cells.	FIG
+39	57	crystal structures	evidence	Ile424–His524 distance measured in the crystal structures correlates with the proliferative effect of triaryl‐ethylene analogs in MCF‐7 cells.	FIG
+102	118	triaryl‐ethylene	chemical	Ile424–His524 distance measured in the crystal structures correlates with the proliferative effect of triaryl‐ethylene analogs in MCF‐7 cells.	FIG
+17	23	Leu354	residue_name_number	In contrast, the Leu354–Leu525 distance correlates with the proliferative effects of OBHS‐N analogs in MCF‐7 cells.	FIG
+24	30	Leu525	residue_name_number	In contrast, the Leu354–Leu525 distance correlates with the proliferative effects of OBHS‐N analogs in MCF‐7 cells.	FIG
+31	39	distance	evidence	In contrast, the Leu354–Leu525 distance correlates with the proliferative effects of OBHS‐N analogs in MCF‐7 cells.	FIG
+85	91	OBHS‐N	chemical	In contrast, the Leu354–Leu525 distance correlates with the proliferative effects of OBHS‐N analogs in MCF‐7 cells.	FIG
+0	24	Structure‐class analysis	experimental_method	Structure‐class analysis of WAY‐C analogs.	FIG
+28	33	WAY‐C	chemical	Structure‐class analysis of WAY‐C analogs.	FIG
+0	5	WAY‐C	chemical	WAY‐C side groups subtly nudge h12 Leu540.	FIG
+31	34	h12	structure_element	WAY‐C side groups subtly nudge h12 Leu540.	FIG
+35	41	Leu540	residue_name_number	WAY‐C side groups subtly nudge h12 Leu540.	FIG
+0	3	ERα	protein	ERα LBD structures bound to 4 distinct WAY‐C analogs were superposed (PDB 4 IU7, 4IV4, 4IVW, 4IW6) (see Datasets EV1 and EV2).	FIG
+4	7	LBD	structure_element	ERα LBD structures bound to 4 distinct WAY‐C analogs were superposed (PDB 4 IU7, 4IV4, 4IVW, 4IW6) (see Datasets EV1 and EV2).	FIG
+8	18	structures	evidence	ERα LBD structures bound to 4 distinct WAY‐C analogs were superposed (PDB 4 IU7, 4IV4, 4IVW, 4IW6) (see Datasets EV1 and EV2).	FIG
+19	27	bound to	protein_state	ERα LBD structures bound to 4 distinct WAY‐C analogs were superposed (PDB 4 IU7, 4IV4, 4IVW, 4IW6) (see Datasets EV1 and EV2).	FIG
+39	44	WAY‐C	chemical	ERα LBD structures bound to 4 distinct WAY‐C analogs were superposed (PDB 4 IU7, 4IV4, 4IVW, 4IW6) (see Datasets EV1 and EV2).	FIG
+58	68	superposed	experimental_method	ERα LBD structures bound to 4 distinct WAY‐C analogs were superposed (PDB 4 IU7, 4IV4, 4IVW, 4IW6) (see Datasets EV1 and EV2).	FIG
+0	24	Structure‐class analysis	experimental_method	Structure‐class analysis of indirect modulators	FIG
+0	24	Structure‐class analysis	experimental_method	Structure‐class analysis of indirect modulators in cluster 1.	FIG
+0	18	Crystal structures	evidence	Crystal structures of the ERα LBD bound to OBHS and OBHS‐N analogs were superposed.	FIG
+26	29	ERα	protein	Crystal structures of the ERα LBD bound to OBHS and OBHS‐N analogs were superposed.	FIG
+30	33	LBD	structure_element	Crystal structures of the ERα LBD bound to OBHS and OBHS‐N analogs were superposed.	FIG
+34	42	bound to	protein_state	Crystal structures of the ERα LBD bound to OBHS and OBHS‐N analogs were superposed.	FIG
+43	47	OBHS	chemical	Crystal structures of the ERα LBD bound to OBHS and OBHS‐N analogs were superposed.	FIG
+52	58	OBHS‐N	chemical	Crystal structures of the ERα LBD bound to OBHS and OBHS‐N analogs were superposed.	FIG
+72	82	superposed	experimental_method	Crystal structures of the ERα LBD bound to OBHS and OBHS‐N analogs were superposed.	FIG
+70	73	h11	structure_element	Arrows indicate chemical variance in the orientation of the different h11‐directed ligand side groups.	FIG
+88	91	h11	structure_element	Panel (B) shows the ligand‐induced conformational variation at the C‐terminal region of h11 (OBHS: PDB 4ZN9, 4ZNH, 4ZNS, 4ZNT, 4ZNU, 4ZNV, and 4ZNW; OBHS‐N: PDB 4ZUB, 4ZUC, 4ZWH, 4ZWK, 5BNU, 5BP6, 5BPR, and 5BQ4).	FIG
+93	97	OBHS	chemical	Panel (B) shows the ligand‐induced conformational variation at the C‐terminal region of h11 (OBHS: PDB 4ZN9, 4ZNH, 4ZNS, 4ZNT, 4ZNU, 4ZNV, and 4ZNW; OBHS‐N: PDB 4ZUB, 4ZUC, 4ZWH, 4ZWK, 5BNU, 5BP6, 5BPR, and 5BQ4).	FIG
+149	155	OBHS‐N	chemical	Panel (B) shows the ligand‐induced conformational variation at the C‐terminal region of h11 (OBHS: PDB 4ZN9, 4ZNH, 4ZNS, 4ZNT, 4ZNU, 4ZNV, and 4ZNW; OBHS‐N: PDB 4ZUB, 4ZUC, 4ZWH, 4ZWK, 5BNU, 5BP6, 5BPR, and 5BQ4).	FIG
+0	24	Structure‐class analysis	experimental_method	Structure‐class analysis of indirect modulators in clusters 2 and 3.	FIG
+0	18	Crystal structures	evidence	Crystal structures of the ERα LBD bound to ligands with cell‐specific activities were superposed.	FIG
+26	29	ERα	protein	Crystal structures of the ERα LBD bound to ligands with cell‐specific activities were superposed.	FIG
+30	33	LBD	structure_element	Crystal structures of the ERα LBD bound to ligands with cell‐specific activities were superposed.	FIG
+34	42	bound to	protein_state	Crystal structures of the ERα LBD bound to ligands with cell‐specific activities were superposed.	FIG
+86	96	superposed	experimental_method	Crystal structures of the ERα LBD bound to ligands with cell‐specific activities were superposed.	FIG
+108	110	h3	structure_element	The bound ligands are shown, and arrows indicate considerable variation in the orientation of the different h3‐, h8‐, h11‐, or h12‐directed ligand side groups.	FIG
+113	115	h8	structure_element	The bound ligands are shown, and arrows indicate considerable variation in the orientation of the different h3‐, h8‐, h11‐, or h12‐directed ligand side groups.	FIG
+118	121	h11	structure_element	The bound ligands are shown, and arrows indicate considerable variation in the orientation of the different h3‐, h8‐, h11‐, or h12‐directed ligand side groups.	FIG
+127	130	h12	structure_element	The bound ligands are shown, and arrows indicate considerable variation in the orientation of the different h3‐, h8‐, h11‐, or h12‐directed ligand side groups.	FIG
+22	25	LBD	structure_element	As visualized in four LBD structures (Srinivasan et al, 2013), WAY‐C analogs were designed with small substitutions that slightly nudge h12 Leu540, without exiting the ligand‐binding pocket (Fig 5G and H).	RESULTS
+26	36	structures	evidence	As visualized in four LBD structures (Srinivasan et al, 2013), WAY‐C analogs were designed with small substitutions that slightly nudge h12 Leu540, without exiting the ligand‐binding pocket (Fig 5G and H).	RESULTS
+63	68	WAY‐C	chemical	As visualized in four LBD structures (Srinivasan et al, 2013), WAY‐C analogs were designed with small substitutions that slightly nudge h12 Leu540, without exiting the ligand‐binding pocket (Fig 5G and H).	RESULTS
+136	139	h12	structure_element	As visualized in four LBD structures (Srinivasan et al, 2013), WAY‐C analogs were designed with small substitutions that slightly nudge h12 Leu540, without exiting the ligand‐binding pocket (Fig 5G and H).	RESULTS
+140	146	Leu540	residue_name_number	As visualized in four LBD structures (Srinivasan et al, 2013), WAY‐C analogs were designed with small substitutions that slightly nudge h12 Leu540, without exiting the ligand‐binding pocket (Fig 5G and H).	RESULTS
+168	189	ligand‐binding pocket	site	As visualized in four LBD structures (Srinivasan et al, 2013), WAY‐C analogs were designed with small substitutions that slightly nudge h12 Leu540, without exiting the ligand‐binding pocket (Fig 5G and H).	RESULTS
+20	23	h12	structure_element	Therefore, changing h12 dynamics maintains the canonical signaling pathway defined by E2 (Fig 1D) to support AF‐2‐driven signaling and recruit NCOA1/2/3 for GREB1‐stimulated proliferation.	RESULTS
+86	88	E2	chemical	Therefore, changing h12 dynamics maintains the canonical signaling pathway defined by E2 (Fig 1D) to support AF‐2‐driven signaling and recruit NCOA1/2/3 for GREB1‐stimulated proliferation.	RESULTS
+109	113	AF‐2	structure_element	Therefore, changing h12 dynamics maintains the canonical signaling pathway defined by E2 (Fig 1D) to support AF‐2‐driven signaling and recruit NCOA1/2/3 for GREB1‐stimulated proliferation.	RESULTS
+143	152	NCOA1/2/3	protein	Therefore, changing h12 dynamics maintains the canonical signaling pathway defined by E2 (Fig 1D) to support AF‐2‐driven signaling and recruit NCOA1/2/3 for GREB1‐stimulated proliferation.	RESULTS
+157	162	GREB1	protein	Therefore, changing h12 dynamics maintains the canonical signaling pathway defined by E2 (Fig 1D) to support AF‐2‐driven signaling and recruit NCOA1/2/3 for GREB1‐stimulated proliferation.	RESULTS
+59	71	AF‐2 surface	site	Ligands with cell‐specific activity alter the shape of the AF‐2 surface	RESULTS
+23	32	tamoxifen	chemical	Direct modulators like tamoxifen drive AF‐1‐dependent cell‐specific activity by completely occluding AF‐2, but it is not known how indirect modulators produce cell‐specific ERα activity.	RESULTS
+39	43	AF‐1	structure_element	Direct modulators like tamoxifen drive AF‐1‐dependent cell‐specific activity by completely occluding AF‐2, but it is not known how indirect modulators produce cell‐specific ERα activity.	RESULTS
+101	105	AF‐2	structure_element	Direct modulators like tamoxifen drive AF‐1‐dependent cell‐specific activity by completely occluding AF‐2, but it is not known how indirect modulators produce cell‐specific ERα activity.	RESULTS
+173	176	ERα	protein	Direct modulators like tamoxifen drive AF‐1‐dependent cell‐specific activity by completely occluding AF‐2, but it is not known how indirect modulators produce cell‐specific ERα activity.	RESULTS
+34	37	LBD	structure_element	Therefore, we examined another 50 LBD structures containing ligands in clusters 2 and 3.	RESULTS
+38	48	structures	evidence	Therefore, we examined another 50 LBD structures containing ligands in clusters 2 and 3.	RESULTS
+6	16	structures	evidence	These structures demonstrated that cell‐specific activity derived from altering the shape of the AF‐2 surface without an extended side chain.	RESULTS
+97	109	AF‐2 surface	site	These structures demonstrated that cell‐specific activity derived from altering the shape of the AF‐2 surface without an extended side chain.	RESULTS
+165	167	h3	structure_element	Ligands in cluster 2 and cluster 3 showed conformational heterogeneity in parts of the scaffold that were directed toward multiple regions of the receptor including h3, h8, h11, h12, and/or the β‐sheets (Fig EV5C–G).	RESULTS
+169	171	h8	structure_element	Ligands in cluster 2 and cluster 3 showed conformational heterogeneity in parts of the scaffold that were directed toward multiple regions of the receptor including h3, h8, h11, h12, and/or the β‐sheets (Fig EV5C–G).	RESULTS
+173	176	h11	structure_element	Ligands in cluster 2 and cluster 3 showed conformational heterogeneity in parts of the scaffold that were directed toward multiple regions of the receptor including h3, h8, h11, h12, and/or the β‐sheets (Fig EV5C–G).	RESULTS
+178	181	h12	structure_element	Ligands in cluster 2 and cluster 3 showed conformational heterogeneity in parts of the scaffold that were directed toward multiple regions of the receptor including h3, h8, h11, h12, and/or the β‐sheets (Fig EV5C–G).	RESULTS
+194	202	β‐sheets	structure_element	Ligands in cluster 2 and cluster 3 showed conformational heterogeneity in parts of the scaffold that were directed toward multiple regions of the receptor including h3, h8, h11, h12, and/or the β‐sheets (Fig EV5C–G).	RESULTS
+14	22	S‐OBHS‐2	chemical	For instance, S‐OBHS‐2 and S‐OBHS‐3 analogs (Fig 2) had similar ERα activity profiles in the different cell types (Fig EV2A–C), but the 2‐ versus 3‐methyl substituted phenol rings altered the correlated signaling patterns in different cell types (Fig 3B lanes 7 and 12).	RESULTS
+27	35	S‐OBHS‐3	chemical	For instance, S‐OBHS‐2 and S‐OBHS‐3 analogs (Fig 2) had similar ERα activity profiles in the different cell types (Fig EV2A–C), but the 2‐ versus 3‐methyl substituted phenol rings altered the correlated signaling patterns in different cell types (Fig 3B lanes 7 and 12).	RESULTS
+64	67	ERα	protein	For instance, S‐OBHS‐2 and S‐OBHS‐3 analogs (Fig 2) had similar ERα activity profiles in the different cell types (Fig EV2A–C), but the 2‐ versus 3‐methyl substituted phenol rings altered the correlated signaling patterns in different cell types (Fig 3B lanes 7 and 12).	RESULTS
+50	62	AF‐2 surface	site	This difference in ligand positioning altered the AF‐2 surface via a shift in the N‐terminus of h12, which directly contacts the coactivator.	RESULTS
+96	99	h12	structure_element	This difference in ligand positioning altered the AF‐2 surface via a shift in the N‐terminus of h12, which directly contacts the coactivator.	RESULTS
+35	44	structure	evidence	This effect is evident in a single structure due to its 1 Å magnitude (Fig 6A and B).	RESULTS
+14	17	h12	structure_element	The shifts in h12 residues Asp538 and Leu539 led to rotation of the coactivator peptide (Fig 6C).	RESULTS
+27	33	Asp538	residue_name_number	The shifts in h12 residues Asp538 and Leu539 led to rotation of the coactivator peptide (Fig 6C).	RESULTS
+38	44	Leu539	residue_name_number	The shifts in h12 residues Asp538 and Leu539 led to rotation of the coactivator peptide (Fig 6C).	RESULTS
+63	75	AF‐2 surface	site	Thus, cell‐specific activity can stem from perturbation of the AF‐2 surface without an extended side chain, which presumably alters the receptor–coregulator interaction profile.	RESULTS
+0	10	S‐OBHS‐2/3	chemical	S‐OBHS‐2/3 analogs subtly distort the AF‐2 surface.	FIG
+38	50	AF‐2 surface	site	S‐OBHS‐2/3 analogs subtly distort the AF‐2 surface.	FIG
+20	37	crystal structure	evidence	Panel (A) shows the crystal structure of an S‐OBHS‐3‐bound ERα LBD (PDB 5DUH).	FIG
+44	58	S‐OBHS‐3‐bound	protein_state	Panel (A) shows the crystal structure of an S‐OBHS‐3‐bound ERα LBD (PDB 5DUH).	FIG
+59	62	ERα	protein	Panel (A) shows the crystal structure of an S‐OBHS‐3‐bound ERα LBD (PDB 5DUH).	FIG
+63	66	LBD	structure_element	Panel (A) shows the crystal structure of an S‐OBHS‐3‐bound ERα LBD (PDB 5DUH).	FIG
+4	20	h3–h12 interface	site	The h3–h12 interface (circled) at AF‐2 (pink) was expanded in panels (B, C).	FIG
+34	38	AF‐2	structure_element	The h3–h12 interface (circled) at AF‐2 (pink) was expanded in panels (B, C).	FIG
+4	20	S‐OBHS‐2/3‐bound	protein_state	The S‐OBHS‐2/3‐bound ERα LBDs were superposed to show shifts in h3 (panel B) and the NCOA2 peptide docked at the AF‐2 surface (panel C).	FIG
+21	24	ERα	protein	The S‐OBHS‐2/3‐bound ERα LBDs were superposed to show shifts in h3 (panel B) and the NCOA2 peptide docked at the AF‐2 surface (panel C).	FIG
+25	29	LBDs	structure_element	The S‐OBHS‐2/3‐bound ERα LBDs were superposed to show shifts in h3 (panel B) and the NCOA2 peptide docked at the AF‐2 surface (panel C).	FIG
+35	45	superposed	experimental_method	The S‐OBHS‐2/3‐bound ERα LBDs were superposed to show shifts in h3 (panel B) and the NCOA2 peptide docked at the AF‐2 surface (panel C).	FIG
+64	66	h3	structure_element	The S‐OBHS‐2/3‐bound ERα LBDs were superposed to show shifts in h3 (panel B) and the NCOA2 peptide docked at the AF‐2 surface (panel C).	FIG
+85	90	NCOA2	protein	The S‐OBHS‐2/3‐bound ERα LBDs were superposed to show shifts in h3 (panel B) and the NCOA2 peptide docked at the AF‐2 surface (panel C).	FIG
+113	125	AF‐2 surface	site	The S‐OBHS‐2/3‐bound ERα LBDs were superposed to show shifts in h3 (panel B) and the NCOA2 peptide docked at the AF‐2 surface (panel C).	FIG
+0	18	Crystal structures	evidence	Crystal structures show that 2,5‐DTP analogs shift h3 and h11 further apart compared to an A‐CD‐ring estrogen (PDB 4PPS, 5DRM, 5DRJ).	FIG
+29	36	2,5‐DTP	chemical	Crystal structures show that 2,5‐DTP analogs shift h3 and h11 further apart compared to an A‐CD‐ring estrogen (PDB 4PPS, 5DRM, 5DRJ).	FIG
+51	53	h3	structure_element	Crystal structures show that 2,5‐DTP analogs shift h3 and h11 further apart compared to an A‐CD‐ring estrogen (PDB 4PPS, 5DRM, 5DRJ).	FIG
+58	61	h11	structure_element	Crystal structures show that 2,5‐DTP analogs shift h3 and h11 further apart compared to an A‐CD‐ring estrogen (PDB 4PPS, 5DRM, 5DRJ).	FIG
+101	109	estrogen	chemical	Crystal structures show that 2,5‐DTP analogs shift h3 and h11 further apart compared to an A‐CD‐ring estrogen (PDB 4PPS, 5DRM, 5DRJ).	FIG
+66	79	2,5‐DTP‐bound	protein_state	The 2F o‐F c electron density map and F o‐F c difference map of a 2,5‐DTP‐bound structure (PDB 5DRJ) were contoured at 1.0 sigma and ± 3.0 sigma, respectively.	FIG
+80	89	structure	evidence	The 2F o‐F c electron density map and F o‐F c difference map of a 2,5‐DTP‐bound structure (PDB 5DRJ) were contoured at 1.0 sigma and ± 3.0 sigma, respectively.	FIG
+21	38	α‐carbon distance	evidence	Average (mean + SEM) α‐carbon distance measured from h3 Thr347 to h11 Leu525 of A‐CD‐, 2,5‐DTP‐, and 3,4‐DTPD‐bound ERα LBDs.	FIG
+53	55	h3	structure_element	Average (mean + SEM) α‐carbon distance measured from h3 Thr347 to h11 Leu525 of A‐CD‐, 2,5‐DTP‐, and 3,4‐DTPD‐bound ERα LBDs.	FIG
+56	62	Thr347	residue_name_number	Average (mean + SEM) α‐carbon distance measured from h3 Thr347 to h11 Leu525 of A‐CD‐, 2,5‐DTP‐, and 3,4‐DTPD‐bound ERα LBDs.	FIG
+66	69	h11	structure_element	Average (mean + SEM) α‐carbon distance measured from h3 Thr347 to h11 Leu525 of A‐CD‐, 2,5‐DTP‐, and 3,4‐DTPD‐bound ERα LBDs.	FIG
+70	76	Leu525	residue_name_number	Average (mean + SEM) α‐carbon distance measured from h3 Thr347 to h11 Leu525 of A‐CD‐, 2,5‐DTP‐, and 3,4‐DTPD‐bound ERα LBDs.	FIG
+80	115	A‐CD‐, 2,5‐DTP‐, and 3,4‐DTPD‐bound	protein_state	Average (mean + SEM) α‐carbon distance measured from h3 Thr347 to h11 Leu525 of A‐CD‐, 2,5‐DTP‐, and 3,4‐DTPD‐bound ERα LBDs.	FIG
+116	119	ERα	protein	Average (mean + SEM) α‐carbon distance measured from h3 Thr347 to h11 Leu525 of A‐CD‐, 2,5‐DTP‐, and 3,4‐DTPD‐bound ERα LBDs.	FIG
+120	124	LBDs	structure_element	Average (mean + SEM) α‐carbon distance measured from h3 Thr347 to h11 Leu525 of A‐CD‐, 2,5‐DTP‐, and 3,4‐DTPD‐bound ERα LBDs.	FIG
+12	28	Student's t‐test	experimental_method	*Two‐tailed Student's t‐test, P = 0.002 (PDB A‐CD: 5DI7, 5DID, 5DIE, 5DIG, and 4PPS; 2,5‐DTP: 4IWC, 5DRM, and 5DRJ; 3,4‐DTPD: 5DTV and 5DU5).	FIG
+30	31	P	evidence	*Two‐tailed Student's t‐test, P = 0.002 (PDB A‐CD: 5DI7, 5DID, 5DIE, 5DIG, and 4PPS; 2,5‐DTP: 4IWC, 5DRM, and 5DRJ; 3,4‐DTPD: 5DTV and 5DU5).	FIG
+45	49	A‐CD	chemical	*Two‐tailed Student's t‐test, P = 0.002 (PDB A‐CD: 5DI7, 5DID, 5DIE, 5DIG, and 4PPS; 2,5‐DTP: 4IWC, 5DRM, and 5DRJ; 3,4‐DTPD: 5DTV and 5DU5).	FIG
+85	92	2,5‐DTP	chemical	*Two‐tailed Student's t‐test, P = 0.002 (PDB A‐CD: 5DI7, 5DID, 5DIE, 5DIG, and 4PPS; 2,5‐DTP: 4IWC, 5DRM, and 5DRJ; 3,4‐DTPD: 5DTV and 5DU5).	FIG
+116	124	3,4‐DTPD	chemical	*Two‐tailed Student's t‐test, P = 0.002 (PDB A‐CD: 5DI7, 5DID, 5DIE, 5DIG, and 4PPS; 2,5‐DTP: 4IWC, 5DRM, and 5DRJ; 3,4‐DTPD: 5DTV and 5DU5).	FIG
+0	18	Crystal structures	evidence	Crystal structures show that a 3,4‐DTPD analog shifts h3 (F) and the NCOA2 (G) peptide compared to an A‐CD‐ring estrogen (PDB 4PPS, 5DTV).	FIG
+31	39	3,4‐DTPD	chemical	Crystal structures show that a 3,4‐DTPD analog shifts h3 (F) and the NCOA2 (G) peptide compared to an A‐CD‐ring estrogen (PDB 4PPS, 5DTV).	FIG
+54	56	h3	structure_element	Crystal structures show that a 3,4‐DTPD analog shifts h3 (F) and the NCOA2 (G) peptide compared to an A‐CD‐ring estrogen (PDB 4PPS, 5DTV).	FIG
+58	59	F	structure_element	Crystal structures show that a 3,4‐DTPD analog shifts h3 (F) and the NCOA2 (G) peptide compared to an A‐CD‐ring estrogen (PDB 4PPS, 5DTV).	FIG
+69	74	NCOA2	protein	Crystal structures show that a 3,4‐DTPD analog shifts h3 (F) and the NCOA2 (G) peptide compared to an A‐CD‐ring estrogen (PDB 4PPS, 5DTV).	FIG
+102	106	A‐CD	chemical	Crystal structures show that a 3,4‐DTPD analog shifts h3 (F) and the NCOA2 (G) peptide compared to an A‐CD‐ring estrogen (PDB 4PPS, 5DTV).	FIG
+112	120	estrogen	chemical	Crystal structures show that a 3,4‐DTPD analog shifts h3 (F) and the NCOA2 (G) peptide compared to an A‐CD‐ring estrogen (PDB 4PPS, 5DTV).	FIG
+0	23	Hierarchical clustering	experimental_method	Hierarchical clustering of ligand‐specific binding of 154 interacting peptides to the ERα LBD was performed in triplicate by MARCoNI analysis.	FIG
+86	89	ERα	protein	Hierarchical clustering of ligand‐specific binding of 154 interacting peptides to the ERα LBD was performed in triplicate by MARCoNI analysis.	FIG
+90	93	LBD	structure_element	Hierarchical clustering of ligand‐specific binding of 154 interacting peptides to the ERα LBD was performed in triplicate by MARCoNI analysis.	FIG
+125	141	MARCoNI analysis	experimental_method	Hierarchical clustering of ligand‐specific binding of 154 interacting peptides to the ERα LBD was performed in triplicate by MARCoNI analysis.	FIG
+4	11	2,5‐DTP	chemical	The 2,5‐DTP analogs showed perturbation of h11, as well as h3, which forms part of the AF‐2 surface.	RESULTS
+43	46	h11	structure_element	The 2,5‐DTP analogs showed perturbation of h11, as well as h3, which forms part of the AF‐2 surface.	RESULTS
+59	61	h3	structure_element	The 2,5‐DTP analogs showed perturbation of h11, as well as h3, which forms part of the AF‐2 surface.	RESULTS
+87	99	AF‐2 surface	site	The 2,5‐DTP analogs showed perturbation of h11, as well as h3, which forms part of the AF‐2 surface.	RESULTS
+25	28	LBD	structure_element	These compounds bind the LBD in an unusual fashion because they have a phenol‐to‐phenol length of ~12 Å, which is longer than steroids and other prototypical ERα agonists that are ~10 Å in length.	RESULTS
+158	161	ERα	protein	These compounds bind the LBD in an unusual fashion because they have a phenol‐to‐phenol length of ~12 Å, which is longer than steroids and other prototypical ERα agonists that are ~10 Å in length.	RESULTS
+33	35	h3	structure_element	One phenol pushed further toward h3 (Fig 6D), while the other phenol pushed toward the C‐terminus of h11 to a greater extent than A‐CD‐ring estrogens (Nwachukwu et al, 2014), which are close structural analogs of E2 that lack a B‐ring (Fig 2).	RESULTS
+101	104	h11	structure_element	One phenol pushed further toward h3 (Fig 6D), while the other phenol pushed toward the C‐terminus of h11 to a greater extent than A‐CD‐ring estrogens (Nwachukwu et al, 2014), which are close structural analogs of E2 that lack a B‐ring (Fig 2).	RESULTS
+130	134	A‐CD	chemical	One phenol pushed further toward h3 (Fig 6D), while the other phenol pushed toward the C‐terminus of h11 to a greater extent than A‐CD‐ring estrogens (Nwachukwu et al, 2014), which are close structural analogs of E2 that lack a B‐ring (Fig 2).	RESULTS
+140	149	estrogens	chemical	One phenol pushed further toward h3 (Fig 6D), while the other phenol pushed toward the C‐terminus of h11 to a greater extent than A‐CD‐ring estrogens (Nwachukwu et al, 2014), which are close structural analogs of E2 that lack a B‐ring (Fig 2).	RESULTS
+213	215	E2	chemical	One phenol pushed further toward h3 (Fig 6D), while the other phenol pushed toward the C‐terminus of h11 to a greater extent than A‐CD‐ring estrogens (Nwachukwu et al, 2014), which are close structural analogs of E2 that lack a B‐ring (Fig 2).	RESULTS
+45	53	distance	evidence	To quantify this difference, we compared the distance between α‐carbons at h3 Thr347 and h11 Leu525 in the set of structures containing 2,5‐DTP analogs (n = 3) or A‐CD‐ring analogs (n = 5) (Fig 6E).	RESULTS
+75	77	h3	structure_element	To quantify this difference, we compared the distance between α‐carbons at h3 Thr347 and h11 Leu525 in the set of structures containing 2,5‐DTP analogs (n = 3) or A‐CD‐ring analogs (n = 5) (Fig 6E).	RESULTS
+78	84	Thr347	residue_name_number	To quantify this difference, we compared the distance between α‐carbons at h3 Thr347 and h11 Leu525 in the set of structures containing 2,5‐DTP analogs (n = 3) or A‐CD‐ring analogs (n = 5) (Fig 6E).	RESULTS
+89	92	h11	structure_element	To quantify this difference, we compared the distance between α‐carbons at h3 Thr347 and h11 Leu525 in the set of structures containing 2,5‐DTP analogs (n = 3) or A‐CD‐ring analogs (n = 5) (Fig 6E).	RESULTS
+93	99	Leu525	residue_name_number	To quantify this difference, we compared the distance between α‐carbons at h3 Thr347 and h11 Leu525 in the set of structures containing 2,5‐DTP analogs (n = 3) or A‐CD‐ring analogs (n = 5) (Fig 6E).	RESULTS
+114	124	structures	evidence	To quantify this difference, we compared the distance between α‐carbons at h3 Thr347 and h11 Leu525 in the set of structures containing 2,5‐DTP analogs (n = 3) or A‐CD‐ring analogs (n = 5) (Fig 6E).	RESULTS
+136	143	2,5‐DTP	chemical	To quantify this difference, we compared the distance between α‐carbons at h3 Thr347 and h11 Leu525 in the set of structures containing 2,5‐DTP analogs (n = 3) or A‐CD‐ring analogs (n = 5) (Fig 6E).	RESULTS
+163	167	A‐CD	chemical	To quantify this difference, we compared the distance between α‐carbons at h3 Thr347 and h11 Leu525 in the set of structures containing 2,5‐DTP analogs (n = 3) or A‐CD‐ring analogs (n = 5) (Fig 6E).	RESULTS
+67	83	Student's t‐test	experimental_method	We observed a difference of 0.4 Å that was significant (two‐tailed Student's t‐test, P = 0.002) due to the very tight clustering of the 2,5‐DTP‐induced LBD conformation.	RESULTS
+85	86	P	evidence	We observed a difference of 0.4 Å that was significant (two‐tailed Student's t‐test, P = 0.002) due to the very tight clustering of the 2,5‐DTP‐induced LBD conformation.	RESULTS
+136	143	2,5‐DTP	chemical	We observed a difference of 0.4 Å that was significant (two‐tailed Student's t‐test, P = 0.002) due to the very tight clustering of the 2,5‐DTP‐induced LBD conformation.	RESULTS
+152	155	LBD	structure_element	We observed a difference of 0.4 Å that was significant (two‐tailed Student's t‐test, P = 0.002) due to the very tight clustering of the 2,5‐DTP‐induced LBD conformation.	RESULTS
+14	16	h3	structure_element	The shifts in h3 suggest these compounds are positioned to alter coregulator preferences.	RESULTS
+4	11	2,5‐DTP	chemical	The 2,5‐DTP and 3,4‐DTP scaffolds are isomeric, but with aryl groups at obtuse and acute angles, respectively (Fig 2).	RESULTS
+16	23	3,4‐DTP	chemical	The 2,5‐DTP and 3,4‐DTP scaffolds are isomeric, but with aryl groups at obtuse and acute angles, respectively (Fig 2).	RESULTS
+4	21	crystal structure	evidence	The crystal structure of ERα in complex with a 3,4‐DTP is unknown; however, we solved two crystal structures of ERα bound to 3,4‐DTPD analogs and one structure containing a furan ligand—all of which have a 3,4‐diaryl configuration (Fig 2; Datasets EV1 and EV2).	RESULTS
+25	28	ERα	protein	The crystal structure of ERα in complex with a 3,4‐DTP is unknown; however, we solved two crystal structures of ERα bound to 3,4‐DTPD analogs and one structure containing a furan ligand—all of which have a 3,4‐diaryl configuration (Fig 2; Datasets EV1 and EV2).	RESULTS
+29	44	in complex with	protein_state	The crystal structure of ERα in complex with a 3,4‐DTP is unknown; however, we solved two crystal structures of ERα bound to 3,4‐DTPD analogs and one structure containing a furan ligand—all of which have a 3,4‐diaryl configuration (Fig 2; Datasets EV1 and EV2).	RESULTS
+47	54	3,4‐DTP	chemical	The crystal structure of ERα in complex with a 3,4‐DTP is unknown; however, we solved two crystal structures of ERα bound to 3,4‐DTPD analogs and one structure containing a furan ligand—all of which have a 3,4‐diaryl configuration (Fig 2; Datasets EV1 and EV2).	RESULTS
+79	85	solved	experimental_method	The crystal structure of ERα in complex with a 3,4‐DTP is unknown; however, we solved two crystal structures of ERα bound to 3,4‐DTPD analogs and one structure containing a furan ligand—all of which have a 3,4‐diaryl configuration (Fig 2; Datasets EV1 and EV2).	RESULTS
+90	108	crystal structures	evidence	The crystal structure of ERα in complex with a 3,4‐DTP is unknown; however, we solved two crystal structures of ERα bound to 3,4‐DTPD analogs and one structure containing a furan ligand—all of which have a 3,4‐diaryl configuration (Fig 2; Datasets EV1 and EV2).	RESULTS
+112	115	ERα	protein	The crystal structure of ERα in complex with a 3,4‐DTP is unknown; however, we solved two crystal structures of ERα bound to 3,4‐DTPD analogs and one structure containing a furan ligand—all of which have a 3,4‐diaryl configuration (Fig 2; Datasets EV1 and EV2).	RESULTS
+116	124	bound to	protein_state	The crystal structure of ERα in complex with a 3,4‐DTP is unknown; however, we solved two crystal structures of ERα bound to 3,4‐DTPD analogs and one structure containing a furan ligand—all of which have a 3,4‐diaryl configuration (Fig 2; Datasets EV1 and EV2).	RESULTS
+125	133	3,4‐DTPD	chemical	The crystal structure of ERα in complex with a 3,4‐DTP is unknown; however, we solved two crystal structures of ERα bound to 3,4‐DTPD analogs and one structure containing a furan ligand—all of which have a 3,4‐diaryl configuration (Fig 2; Datasets EV1 and EV2).	RESULTS
+150	159	structure	evidence	The crystal structure of ERα in complex with a 3,4‐DTP is unknown; however, we solved two crystal structures of ERα bound to 3,4‐DTPD analogs and one structure containing a furan ligand—all of which have a 3,4‐diaryl configuration (Fig 2; Datasets EV1 and EV2).	RESULTS
+173	178	furan	chemical	The crystal structure of ERα in complex with a 3,4‐DTP is unknown; however, we solved two crystal structures of ERα bound to 3,4‐DTPD analogs and one structure containing a furan ligand—all of which have a 3,4‐diaryl configuration (Fig 2; Datasets EV1 and EV2).	RESULTS
+9	19	structures	evidence	In these structures, the A‐ring mimetic of the 3,4‐DTPD scaffold bound h3 Glu353 as expected, but the other phenol wrapped around h3 to form a hydrogen bond with Thr347, indicating a change in binding epitopes in the ERα ligand‐binding pocket (Fig 6F).	RESULTS
+47	55	3,4‐DTPD	chemical	In these structures, the A‐ring mimetic of the 3,4‐DTPD scaffold bound h3 Glu353 as expected, but the other phenol wrapped around h3 to form a hydrogen bond with Thr347, indicating a change in binding epitopes in the ERα ligand‐binding pocket (Fig 6F).	RESULTS
+71	73	h3	structure_element	In these structures, the A‐ring mimetic of the 3,4‐DTPD scaffold bound h3 Glu353 as expected, but the other phenol wrapped around h3 to form a hydrogen bond with Thr347, indicating a change in binding epitopes in the ERα ligand‐binding pocket (Fig 6F).	RESULTS
+74	80	Glu353	residue_name_number	In these structures, the A‐ring mimetic of the 3,4‐DTPD scaffold bound h3 Glu353 as expected, but the other phenol wrapped around h3 to form a hydrogen bond with Thr347, indicating a change in binding epitopes in the ERα ligand‐binding pocket (Fig 6F).	RESULTS
+130	132	h3	structure_element	In these structures, the A‐ring mimetic of the 3,4‐DTPD scaffold bound h3 Glu353 as expected, but the other phenol wrapped around h3 to form a hydrogen bond with Thr347, indicating a change in binding epitopes in the ERα ligand‐binding pocket (Fig 6F).	RESULTS
+162	168	Thr347	residue_name_number	In these structures, the A‐ring mimetic of the 3,4‐DTPD scaffold bound h3 Glu353 as expected, but the other phenol wrapped around h3 to form a hydrogen bond with Thr347, indicating a change in binding epitopes in the ERα ligand‐binding pocket (Fig 6F).	RESULTS
+217	220	ERα	protein	In these structures, the A‐ring mimetic of the 3,4‐DTPD scaffold bound h3 Glu353 as expected, but the other phenol wrapped around h3 to form a hydrogen bond with Thr347, indicating a change in binding epitopes in the ERα ligand‐binding pocket (Fig 6F).	RESULTS
+221	242	ligand‐binding pocket	site	In these structures, the A‐ring mimetic of the 3,4‐DTPD scaffold bound h3 Glu353 as expected, but the other phenol wrapped around h3 to form a hydrogen bond with Thr347, indicating a change in binding epitopes in the ERα ligand‐binding pocket (Fig 6F).	RESULTS
+4	12	3,4‐DTPD	chemical	The 3,4‐DTPD analogs also induced a shift in h3 positioning, which translated again into a shift in the bound coactivator peptide (Fig 6F).	RESULTS
+45	47	h3	structure_element	The 3,4‐DTPD analogs also induced a shift in h3 positioning, which translated again into a shift in the bound coactivator peptide (Fig 6F).	RESULTS
+48	56	S‐OBHS‐2	chemical	Therefore, these indirect modulators, including S‐OBHS‐2, S‐OBHS‐3, 2,5‐DTP, and 3,4‐DTPD analogs—all of which show cell‐specific activity profiles—induced shifts in h3 and h12 that were transmitted to the coactivator peptide via an altered AF‐2 surface.	RESULTS
+58	66	S‐OBHS‐3	chemical	Therefore, these indirect modulators, including S‐OBHS‐2, S‐OBHS‐3, 2,5‐DTP, and 3,4‐DTPD analogs—all of which show cell‐specific activity profiles—induced shifts in h3 and h12 that were transmitted to the coactivator peptide via an altered AF‐2 surface.	RESULTS
+68	75	2,5‐DTP	chemical	Therefore, these indirect modulators, including S‐OBHS‐2, S‐OBHS‐3, 2,5‐DTP, and 3,4‐DTPD analogs—all of which show cell‐specific activity profiles—induced shifts in h3 and h12 that were transmitted to the coactivator peptide via an altered AF‐2 surface.	RESULTS
+81	89	3,4‐DTPD	chemical	Therefore, these indirect modulators, including S‐OBHS‐2, S‐OBHS‐3, 2,5‐DTP, and 3,4‐DTPD analogs—all of which show cell‐specific activity profiles—induced shifts in h3 and h12 that were transmitted to the coactivator peptide via an altered AF‐2 surface.	RESULTS
+166	168	h3	structure_element	Therefore, these indirect modulators, including S‐OBHS‐2, S‐OBHS‐3, 2,5‐DTP, and 3,4‐DTPD analogs—all of which show cell‐specific activity profiles—induced shifts in h3 and h12 that were transmitted to the coactivator peptide via an altered AF‐2 surface.	RESULTS
+173	176	h12	structure_element	Therefore, these indirect modulators, including S‐OBHS‐2, S‐OBHS‐3, 2,5‐DTP, and 3,4‐DTPD analogs—all of which show cell‐specific activity profiles—induced shifts in h3 and h12 that were transmitted to the coactivator peptide via an altered AF‐2 surface.	RESULTS
+241	253	AF‐2 surface	site	Therefore, these indirect modulators, including S‐OBHS‐2, S‐OBHS‐3, 2,5‐DTP, and 3,4‐DTPD analogs—all of which show cell‐specific activity profiles—induced shifts in h3 and h12 that were transmitted to the coactivator peptide via an altered AF‐2 surface.	RESULTS
+20	32	AF‐2 surface	site	To test whether the AF‐2 surface shows changes in shape in solution, we used the microarray assay for real‐time coregulator–nuclear receptor interaction (MARCoNI) analysis (Aarts et al, 2013).	RESULTS
+81	152	microarray assay for real‐time coregulator–nuclear receptor interaction	experimental_method	To test whether the AF‐2 surface shows changes in shape in solution, we used the microarray assay for real‐time coregulator–nuclear receptor interaction (MARCoNI) analysis (Aarts et al, 2013).	RESULTS
+154	161	MARCoNI	experimental_method	To test whether the AF‐2 surface shows changes in shape in solution, we used the microarray assay for real‐time coregulator–nuclear receptor interaction (MARCoNI) analysis (Aarts et al, 2013).	RESULTS
+47	50	ERα	protein	Here, the ligand‐dependent interactions of the ERα LBD with over 150 distinct LxxLL motif peptides were assayed to define structural fingerprints for the AF‐2 surface, in a manner similar to the use of phage display peptides as structural probes (Connor et al, 2001).	RESULTS
+51	54	LBD	structure_element	Here, the ligand‐dependent interactions of the ERα LBD with over 150 distinct LxxLL motif peptides were assayed to define structural fingerprints for the AF‐2 surface, in a manner similar to the use of phage display peptides as structural probes (Connor et al, 2001).	RESULTS
+78	89	LxxLL motif	structure_element	Here, the ligand‐dependent interactions of the ERα LBD with over 150 distinct LxxLL motif peptides were assayed to define structural fingerprints for the AF‐2 surface, in a manner similar to the use of phage display peptides as structural probes (Connor et al, 2001).	RESULTS
+154	166	AF‐2 surface	site	Here, the ligand‐dependent interactions of the ERα LBD with over 150 distinct LxxLL motif peptides were assayed to define structural fingerprints for the AF‐2 surface, in a manner similar to the use of phage display peptides as structural probes (Connor et al, 2001).	RESULTS
+202	224	phage display peptides	experimental_method	Here, the ligand‐dependent interactions of the ERα LBD with over 150 distinct LxxLL motif peptides were assayed to define structural fingerprints for the AF‐2 surface, in a manner similar to the use of phage display peptides as structural probes (Connor et al, 2001).	RESULTS
+79	86	2,5‐DTP	chemical	Despite the similar average activities of these ligand classes (Fig 3A and B), 2,5‐DTP and 3,4‐DTP analogs displayed remarkably different peptide recruitment patterns (Fig 6H), consistent with the structural analyses.	RESULTS
+91	98	3,4‐DTP	chemical	Despite the similar average activities of these ligand classes (Fig 3A and B), 2,5‐DTP and 3,4‐DTP analogs displayed remarkably different peptide recruitment patterns (Fig 6H), consistent with the structural analyses.	RESULTS
+197	216	structural analyses	experimental_method	Despite the similar average activities of these ligand classes (Fig 3A and B), 2,5‐DTP and 3,4‐DTP analogs displayed remarkably different peptide recruitment patterns (Fig 6H), consistent with the structural analyses.	RESULTS
+0	23	Hierarchical clustering	experimental_method	Hierarchical clustering revealed that many of the 2,5‐DTP analogs recapitulated most of the peptide recruitment and dismissal patterns observed with E2 (Fig 6H).	RESULTS
+50	57	2,5‐DTP	chemical	Hierarchical clustering revealed that many of the 2,5‐DTP analogs recapitulated most of the peptide recruitment and dismissal patterns observed with E2 (Fig 6H).	RESULTS
+149	151	E2	chemical	Hierarchical clustering revealed that many of the 2,5‐DTP analogs recapitulated most of the peptide recruitment and dismissal patterns observed with E2 (Fig 6H).	RESULTS
+71	73	E2	chemical	However, there was a unique cluster of peptides that were recruited by E2 but not the 2,5‐DTP analogs.	RESULTS
+86	93	2,5‐DTP	chemical	However, there was a unique cluster of peptides that were recruited by E2 but not the 2,5‐DTP analogs.	RESULTS
+13	20	3,4‐DTP	chemical	In contrast, 3,4‐DTP analogs dismissed most of the peptides from the AF‐2 surface (Fig 6H).	RESULTS
+69	81	AF‐2 surface	site	In contrast, 3,4‐DTP analogs dismissed most of the peptides from the AF‐2 surface (Fig 6H).	RESULTS
+54	63	thiophene	chemical	Thus, the isomeric attachment of diaryl groups to the thiophene core changed the AF‐2 surface from inside the ligand‐binding pocket, as predicted by the crystal structures.	RESULTS
+81	93	AF‐2 surface	site	Thus, the isomeric attachment of diaryl groups to the thiophene core changed the AF‐2 surface from inside the ligand‐binding pocket, as predicted by the crystal structures.	RESULTS
+110	131	ligand‐binding pocket	site	Thus, the isomeric attachment of diaryl groups to the thiophene core changed the AF‐2 surface from inside the ligand‐binding pocket, as predicted by the crystal structures.	RESULTS
+153	171	crystal structures	evidence	Thus, the isomeric attachment of diaryl groups to the thiophene core changed the AF‐2 surface from inside the ligand‐binding pocket, as predicted by the crystal structures.	RESULTS
+198	210	AF‐2 surface	site	Together, these findings suggest that without an extended side chain, cell‐specific activity stems from different coregulator recruitment profiles, due to unique ligand‐induced conformations of the AF‐2 surface, in addition to differential usage of AF‐1.	RESULTS
+249	253	AF‐1	structure_element	Together, these findings suggest that without an extended side chain, cell‐specific activity stems from different coregulator recruitment profiles, due to unique ligand‐induced conformations of the AF‐2 surface, in addition to differential usage of AF‐1.	RESULTS
+62	79	h11–h12 interface	site	Indirect modulators in cluster 1 avoid this by perturbing the h11–h12 interface, and modulating the dynamics of h12 without changing the shape of AF‐2 when stabilized.	RESULTS
+112	115	h12	structure_element	Indirect modulators in cluster 1 avoid this by perturbing the h11–h12 interface, and modulating the dynamics of h12 without changing the shape of AF‐2 when stabilized.	RESULTS
+146	150	AF‐2	structure_element	Indirect modulators in cluster 1 avoid this by perturbing the h11–h12 interface, and modulating the dynamics of h12 without changing the shape of AF‐2 when stabilized.	RESULTS
+106	109	ERα	protein	Our goal was to identify a minimal set of predictors that would link specific structural perturbations to ERα signaling pathways that control cell‐specific signaling and proliferation.	DISCUSS
+148	151	h12	structure_element	We found a very strong set of predictors, where ligands in cluster 1, defined by similar signaling across cell types, showed indirect modulation of h12 dynamics via the h11–12 interface or slight contact with h12.	DISCUSS
+169	185	h11–12 interface	site	We found a very strong set of predictors, where ligands in cluster 1, defined by similar signaling across cell types, showed indirect modulation of h12 dynamics via the h11–12 interface or slight contact with h12.	DISCUSS
+209	212	h12	structure_element	We found a very strong set of predictors, where ligands in cluster 1, defined by similar signaling across cell types, showed indirect modulation of h12 dynamics via the h11–12 interface or slight contact with h12.	DISCUSS
+73	77	AF‐2	structure_element	This perturbation determined proliferation that correlated strongly with AF‐2 activity, recruitment of NCOA1/2/3 family members, and induction of the GREB1 gene, consistent with the canonical ERα signaling pathway (Fig 1D).	DISCUSS
+103	112	NCOA1/2/3	protein	This perturbation determined proliferation that correlated strongly with AF‐2 activity, recruitment of NCOA1/2/3 family members, and induction of the GREB1 gene, consistent with the canonical ERα signaling pathway (Fig 1D).	DISCUSS
+150	155	GREB1	protein	This perturbation determined proliferation that correlated strongly with AF‐2 activity, recruitment of NCOA1/2/3 family members, and induction of the GREB1 gene, consistent with the canonical ERα signaling pathway (Fig 1D).	DISCUSS
+192	195	ERα	protein	This perturbation determined proliferation that correlated strongly with AF‐2 activity, recruitment of NCOA1/2/3 family members, and induction of the GREB1 gene, consistent with the canonical ERα signaling pathway (Fig 1D).	DISCUSS
+26	34	deletion	experimental_method	For ligands in cluster 1, deletion of AF‐1 reduced activity to varying degrees, but did not change the underlying signaling patterns established through AF‐2.	DISCUSS
+38	42	AF‐1	structure_element	For ligands in cluster 1, deletion of AF‐1 reduced activity to varying degrees, but did not change the underlying signaling patterns established through AF‐2.	DISCUSS
+153	157	AF‐2	structure_element	For ligands in cluster 1, deletion of AF‐1 reduced activity to varying degrees, but did not change the underlying signaling patterns established through AF‐2.	DISCUSS
+68	71	h12	structure_element	In contrast, an extended side chain designed to directly reposition h12 and completely disrupt the AF‐2 surface results in cell‐specific signaling.	DISCUSS
+99	111	AF‐2 surface	site	In contrast, an extended side chain designed to directly reposition h12 and completely disrupt the AF‐2 surface results in cell‐specific signaling.	DISCUSS
+137	140	LBD	structure_element	Compared to cluster 1, the structural rules are less clear in clusters 2 and 3, but a number of indirect modulator classes perturbed the LBD conformation at the intersection of h3, the h12 N‐terminus, and the AF‐2 surface.	DISCUSS
+177	179	h3	structure_element	Compared to cluster 1, the structural rules are less clear in clusters 2 and 3, but a number of indirect modulator classes perturbed the LBD conformation at the intersection of h3, the h12 N‐terminus, and the AF‐2 surface.	DISCUSS
+185	188	h12	structure_element	Compared to cluster 1, the structural rules are less clear in clusters 2 and 3, but a number of indirect modulator classes perturbed the LBD conformation at the intersection of h3, the h12 N‐terminus, and the AF‐2 surface.	DISCUSS
+209	221	AF‐2 surface	site	Compared to cluster 1, the structural rules are less clear in clusters 2 and 3, but a number of indirect modulator classes perturbed the LBD conformation at the intersection of h3, the h12 N‐terminus, and the AF‐2 surface.	DISCUSS
+46	50	AF‐2	structure_element	Ligands in these classes altered the shape of AF‐2 to affect coregulator preferences.	DISCUSS
+68	71	ERα	protein	For direct and indirect modulators in cluster 2 or 3, the canonical ERα signaling pathway involving recruitment of NCOA1/2/3 and induction of GREB1 did not generally predict their proliferative effects, indicating an alternate causal model (Fig 1E).	DISCUSS
+115	124	NCOA1/2/3	protein	For direct and indirect modulators in cluster 2 or 3, the canonical ERα signaling pathway involving recruitment of NCOA1/2/3 and induction of GREB1 did not generally predict their proliferative effects, indicating an alternate causal model (Fig 1E).	DISCUSS
+142	147	GREB1	protein	For direct and indirect modulators in cluster 2 or 3, the canonical ERα signaling pathway involving recruitment of NCOA1/2/3 and induction of GREB1 did not generally predict their proliferative effects, indicating an alternate causal model (Fig 1E).	DISCUSS
+71	96	ligand–receptor interface	site	These principles outlined above provide a structural basis for how the ligand–receptor interface leads to different signaling specificities through AF‐1 and AF‐2.	DISCUSS
+148	152	AF‐1	structure_element	These principles outlined above provide a structural basis for how the ligand–receptor interface leads to different signaling specificities through AF‐1 and AF‐2.	DISCUSS
+157	161	AF‐2	structure_element	These principles outlined above provide a structural basis for how the ligand–receptor interface leads to different signaling specificities through AF‐1 and AF‐2.	DISCUSS
+36	39	h12	structure_element	It is noteworthy that regulation of h12 dynamics indirectly through h11 can virtually abolish AF‐2 activity, and yet still drive robust transcriptional activity through AF‐1, as demonstrated with the OBHS series.	DISCUSS
+68	71	h11	structure_element	It is noteworthy that regulation of h12 dynamics indirectly through h11 can virtually abolish AF‐2 activity, and yet still drive robust transcriptional activity through AF‐1, as demonstrated with the OBHS series.	DISCUSS
+94	98	AF‐2	structure_element	It is noteworthy that regulation of h12 dynamics indirectly through h11 can virtually abolish AF‐2 activity, and yet still drive robust transcriptional activity through AF‐1, as demonstrated with the OBHS series.	DISCUSS
+169	173	AF‐1	structure_element	It is noteworthy that regulation of h12 dynamics indirectly through h11 can virtually abolish AF‐2 activity, and yet still drive robust transcriptional activity through AF‐1, as demonstrated with the OBHS series.	DISCUSS
+200	204	OBHS	chemical	It is noteworthy that regulation of h12 dynamics indirectly through h11 can virtually abolish AF‐2 activity, and yet still drive robust transcriptional activity through AF‐1, as demonstrated with the OBHS series.	DISCUSS
+47	56	NCOA1/2/3	protein	This finding can be explained by the fact that NCOA1/2/3 contain distinct binding sites for interaction with AF‐1 and AF‐2 (McInerney et al, 1996; Webb et al, 1998), which allows ligands to nucleate ERα–NCOA1/2/3 interaction through AF‐2, and reinforce this interaction with additional binding to AF‐1.	DISCUSS
+74	87	binding sites	site	This finding can be explained by the fact that NCOA1/2/3 contain distinct binding sites for interaction with AF‐1 and AF‐2 (McInerney et al, 1996; Webb et al, 1998), which allows ligands to nucleate ERα–NCOA1/2/3 interaction through AF‐2, and reinforce this interaction with additional binding to AF‐1.	DISCUSS
+109	113	AF‐1	structure_element	This finding can be explained by the fact that NCOA1/2/3 contain distinct binding sites for interaction with AF‐1 and AF‐2 (McInerney et al, 1996; Webb et al, 1998), which allows ligands to nucleate ERα–NCOA1/2/3 interaction through AF‐2, and reinforce this interaction with additional binding to AF‐1.	DISCUSS
+118	122	AF‐2	structure_element	This finding can be explained by the fact that NCOA1/2/3 contain distinct binding sites for interaction with AF‐1 and AF‐2 (McInerney et al, 1996; Webb et al, 1998), which allows ligands to nucleate ERα–NCOA1/2/3 interaction through AF‐2, and reinforce this interaction with additional binding to AF‐1.	DISCUSS
+199	212	ERα–NCOA1/2/3	complex_assembly	This finding can be explained by the fact that NCOA1/2/3 contain distinct binding sites for interaction with AF‐1 and AF‐2 (McInerney et al, 1996; Webb et al, 1998), which allows ligands to nucleate ERα–NCOA1/2/3 interaction through AF‐2, and reinforce this interaction with additional binding to AF‐1.	DISCUSS
+233	237	AF‐2	structure_element	This finding can be explained by the fact that NCOA1/2/3 contain distinct binding sites for interaction with AF‐1 and AF‐2 (McInerney et al, 1996; Webb et al, 1998), which allows ligands to nucleate ERα–NCOA1/2/3 interaction through AF‐2, and reinforce this interaction with additional binding to AF‐1.	DISCUSS
+297	301	AF‐1	structure_element	This finding can be explained by the fact that NCOA1/2/3 contain distinct binding sites for interaction with AF‐1 and AF‐2 (McInerney et al, 1996; Webb et al, 1998), which allows ligands to nucleate ERα–NCOA1/2/3 interaction through AF‐2, and reinforce this interaction with additional binding to AF‐1.	DISCUSS
+20	24	AF‐2	structure_element	Completely blocking AF‐2 with an extended side chain or altering the shape of AF‐2 changes the preference away from NCOA1/2/3 for determining GREB1 levels and proliferation of breast cancer cells.	DISCUSS
+78	82	AF‐2	structure_element	Completely blocking AF‐2 with an extended side chain or altering the shape of AF‐2 changes the preference away from NCOA1/2/3 for determining GREB1 levels and proliferation of breast cancer cells.	DISCUSS
+116	125	NCOA1/2/3	protein	Completely blocking AF‐2 with an extended side chain or altering the shape of AF‐2 changes the preference away from NCOA1/2/3 for determining GREB1 levels and proliferation of breast cancer cells.	DISCUSS
+142	147	GREB1	protein	Completely blocking AF‐2 with an extended side chain or altering the shape of AF‐2 changes the preference away from NCOA1/2/3 for determining GREB1 levels and proliferation of breast cancer cells.	DISCUSS
+0	4	AF‐2	structure_element	AF‐2 blockade also allows AF‐1 to function independently, which is important since AF‐1 drives tissue‐selective effects in vivo.	DISCUSS
+26	30	AF‐1	structure_element	AF‐2 blockade also allows AF‐1 to function independently, which is important since AF‐1 drives tissue‐selective effects in vivo.	DISCUSS
+83	87	AF‐1	structure_element	AF‐2 blockade also allows AF‐1 to function independently, which is important since AF‐1 drives tissue‐selective effects in vivo.	DISCUSS
+27	31	AF‐1	structure_element	This was demonstrated with AF‐1 knockout mice that show E2‐dependent vascular protection, but not uterine proliferation, thus highlighting the role of AF‐1 in tissue‐selective or cell‐specific signaling (Billon‐Gales et al, 2009; Abot et al, 2013).	DISCUSS
+56	58	E2	chemical	This was demonstrated with AF‐1 knockout mice that show E2‐dependent vascular protection, but not uterine proliferation, thus highlighting the role of AF‐1 in tissue‐selective or cell‐specific signaling (Billon‐Gales et al, 2009; Abot et al, 2013).	DISCUSS
+151	155	AF‐1	structure_element	This was demonstrated with AF‐1 knockout mice that show E2‐dependent vascular protection, but not uterine proliferation, thus highlighting the role of AF‐1 in tissue‐selective or cell‐specific signaling (Billon‐Gales et al, 2009; Abot et al, 2013).	DISCUSS
+23	26	LBD	structure_element	Here, we examined many LBD structures and tested several variables that were not predictive, including ERβ activity, the strength of AF‐1 signaling, and NCOA3 occupancy at the GREB1 gene.	DISCUSS
+27	37	structures	evidence	Here, we examined many LBD structures and tested several variables that were not predictive, including ERβ activity, the strength of AF‐1 signaling, and NCOA3 occupancy at the GREB1 gene.	DISCUSS
+103	106	ERβ	protein	Here, we examined many LBD structures and tested several variables that were not predictive, including ERβ activity, the strength of AF‐1 signaling, and NCOA3 occupancy at the GREB1 gene.	DISCUSS
+133	137	AF‐1	structure_element	Here, we examined many LBD structures and tested several variables that were not predictive, including ERβ activity, the strength of AF‐1 signaling, and NCOA3 occupancy at the GREB1 gene.	DISCUSS
+153	158	NCOA3	protein	Here, we examined many LBD structures and tested several variables that were not predictive, including ERβ activity, the strength of AF‐1 signaling, and NCOA3 occupancy at the GREB1 gene.	DISCUSS
+176	181	GREB1	protein	Here, we examined many LBD structures and tested several variables that were not predictive, including ERβ activity, the strength of AF‐1 signaling, and NCOA3 occupancy at the GREB1 gene.	DISCUSS
+25	35	structures	evidence	Similarly, we visualized structures to identify patterns.	DISCUSS
+13	26	phage display	experimental_method	For example, phage display was used to identify the androgen receptor interactome, which was cloned into an M2H library and used to identify clusters of ligand‐selective interactions (Norris et al, 2009).	DISCUSS
+108	111	M2H	experimental_method	For example, phage display was used to identify the androgen receptor interactome, which was cloned into an M2H library and used to identify clusters of ligand‐selective interactions (Norris et al, 2009).	DISCUSS
+19	34	siRNA screening	experimental_method	Also, we have used siRNA screening to identify a number of coregulators required for ERα‐mediated repression of the IL‐6 gene (Nwachukwu et al, 2014).	DISCUSS
+85	88	ERα	protein	Also, we have used siRNA screening to identify a number of coregulators required for ERα‐mediated repression of the IL‐6 gene (Nwachukwu et al, 2014).	DISCUSS
+17	47	inter‐atomic distance matrices	evidence	If we calculated inter‐atomic distance matrices containing 4,000 atoms per structure × 76 ligand–receptor complexes, we would have 3 × 105 predictions.	DISCUSS
+19	33	atomic vectors	evidence	We have identified atomic vectors for the OBHS‐N and triaryl‐ethylene classes that predict ligand response (Fig 5E and F).	DISCUSS
+42	48	OBHS‐N	chemical	We have identified atomic vectors for the OBHS‐N and triaryl‐ethylene classes that predict ligand response (Fig 5E and F).	DISCUSS
+53	69	triaryl‐ethylene	chemical	We have identified atomic vectors for the OBHS‐N and triaryl‐ethylene classes that predict ligand response (Fig 5E and F).	DISCUSS
+52	58	OBHS‐N	chemical	Indeed, the most anti‐proliferative compound in the OBHS‐N series had a fulvestrant‐like profile across a battery of assays (S. Srinivasan et al, in preparation).	DISCUSS
+27	32	WAY‐C	chemical	Secondly, our finding that WAY‐C compounds do not rely of AF‐1 for signaling efficacy may derive from the slight contacts with h12 observed in crystal structures (Figs 3B and 5H), unlike other compounds in cluster 1 that dislocate h11 and rely on AF‐1 for signaling efficacy (Figs 3B and 5C, and EV5B).	DISCUSS
+58	62	AF‐1	structure_element	Secondly, our finding that WAY‐C compounds do not rely of AF‐1 for signaling efficacy may derive from the slight contacts with h12 observed in crystal structures (Figs 3B and 5H), unlike other compounds in cluster 1 that dislocate h11 and rely on AF‐1 for signaling efficacy (Figs 3B and 5C, and EV5B).	DISCUSS
+127	130	h12	structure_element	Secondly, our finding that WAY‐C compounds do not rely of AF‐1 for signaling efficacy may derive from the slight contacts with h12 observed in crystal structures (Figs 3B and 5H), unlike other compounds in cluster 1 that dislocate h11 and rely on AF‐1 for signaling efficacy (Figs 3B and 5C, and EV5B).	DISCUSS
+143	161	crystal structures	evidence	Secondly, our finding that WAY‐C compounds do not rely of AF‐1 for signaling efficacy may derive from the slight contacts with h12 observed in crystal structures (Figs 3B and 5H), unlike other compounds in cluster 1 that dislocate h11 and rely on AF‐1 for signaling efficacy (Figs 3B and 5C, and EV5B).	DISCUSS
+231	234	h11	structure_element	Secondly, our finding that WAY‐C compounds do not rely of AF‐1 for signaling efficacy may derive from the slight contacts with h12 observed in crystal structures (Figs 3B and 5H), unlike other compounds in cluster 1 that dislocate h11 and rely on AF‐1 for signaling efficacy (Figs 3B and 5C, and EV5B).	DISCUSS
+247	251	AF‐1	structure_element	Secondly, our finding that WAY‐C compounds do not rely of AF‐1 for signaling efficacy may derive from the slight contacts with h12 observed in crystal structures (Figs 3B and 5H), unlike other compounds in cluster 1 that dislocate h11 and rely on AF‐1 for signaling efficacy (Figs 3B and 5C, and EV5B).	DISCUSS
+47	59	AF‐2 surface	site	Some of these ligands altered the shape of the AF‐2 surface by perturbing the h3–h12 interface, thus providing a route to new SERM‐like activity profiles by combining indirect and direct modulation of receptor structure.	DISCUSS
+78	94	h3–h12 interface	site	Some of these ligands altered the shape of the AF‐2 surface by perturbing the h3–h12 interface, thus providing a route to new SERM‐like activity profiles by combining indirect and direct modulation of receptor structure.	DISCUSS
+156	159	ERα	protein	Incorporation of statistical approaches to understand relationships between structure and signaling variables moves us toward predictive models for complex ERα‐mediated responses such as in vivo uterine proliferation or tumor growth, and more generally toward structure‐based design for other allosteric drug targets including GPCRs and other nuclear receptors.	DISCUSS
+327	332	GPCRs	protein_type	Incorporation of statistical approaches to understand relationships between structure and signaling variables moves us toward predictive models for complex ERα‐mediated responses such as in vivo uterine proliferation or tumor growth, and more generally toward structure‐based design for other allosteric drug targets including GPCRs and other nuclear receptors.	DISCUSS
+343	360	nuclear receptors	protein_type	Incorporation of statistical approaches to understand relationships between structure and signaling variables moves us toward predictive models for complex ERα‐mediated responses such as in vivo uterine proliferation or tumor growth, and more generally toward structure‐based design for other allosteric drug targets including GPCRs and other nuclear receptors.	DISCUSS