",unsafe_allow_html=True)
st.markdown("""---""")
with st.expander("# **:green[What's new?]**"):
st.markdown(what_new)
def ab_sidebar():
st.title('**About**')
st.markdown("""---""")
with st.expander("# **:green[What's new?]**"):
st.markdown(what_new)
# @st.cache_data(experimental_allow_widgets=True)
# if 'sidebar_con' not in st.session_state:
# sidebar_con = st.empty()
# def render_sidebar(page,step):
# sidebar_con.empty()
# with sidebar_con.container():
# if page == 0:
# with st.sidebar():
# oam_sidebar(step)
def colorize_step(n_step,cur_step):
color_list = ['grey']*n_step
for i in range(cur_step):
color_list[i] = 'mediumseagreen'
color_list[cur_step] = 'tomato'
if cur_step == n_step-1:
color_list[cur_step] = 'mediumseagreen'
return color_list
def form_header():
st.markdown("
Molecular Optimization using Junction Tree Variational Autoencoder
",unsafe_allow_html=True)
st.markdown("
Gia-Bao Truong
",unsafe_allow_html=True)
with st.expander(':star2: About the model'):
st.markdown("
Based on Junction Tree Variational Autoencoder for Molecular Graph Generation (JTVAE)
",unsafe_allow_html=True)
st.markdown("
Wengong Jin, Regina Barzilay, Tommi Jaakkola
",unsafe_allow_html=True)
# determines button color which should be red when user is on that given step
oam_type = 'primary' if st.session_state['current_view'] == 0 else 'secondary'
oab_type = 'primary' if st.session_state['current_view'] == 1 else 'secondary'
ab_type = 'primary' if st.session_state['current_view'] == 2 else 'secondary'
step_cols = st.columns([.2,.85,.85,.85,.2])
step_cols[1].button('Optimize a molecule',on_click=set_page_view,args=[0],type=oam_type,use_container_width=True)
step_cols[2].button('Optimize a batch',on_click=set_page_view,args=[1],type=oab_type,use_container_width=True)
step_cols[3].button('About',on_click=set_page_view,args=[2],type=ab_type,use_container_width=True)
st.empty()
def form_body():
body_container = st.empty()
###### Optimize a molecule ######
if st.session_state['current_view'] == 0:
body_container.empty()
with body_container.container():
Optimize_a_molecule()
###### Optimize a batch ######
if st.session_state['current_view'] == 1:
body_container.empty()
with body_container.container():
Optimize_a_batch()
###### About ######
if st.session_state['current_view'] == 2:
body_container.empty()
with body_container.container():
About()
def About():
descrip_model = '''
We seek to automate the design of molecules based on specific chemical properties. In computational terms, this task involves continuous embedding and generation of molecular graphs. Our primary contribution is the direct realization of molecular graphs, a task previously approached by generating linear SMILES strings instead of graphs. Our junction tree variational autoencoder generates molecular graphs in two phases, by first generating a tree-structured scaffold over chemical substructures, and then combining them into a molecule with a graph message passing network. This approach allows us to incrementally expand molecules while maintaining chemical validity at every step. We evaluate our model on multiple tasks ranging from molecular generation to optimization. Across these tasks, our model outperforms previous state-of-the-art baselines by a significant margin.
'''
img_caption = '''
Figure 3. Overview of our method: A molecular graph G is first decomposed into its junction tree TG, where each colored node in the tree represents a substructure in the molecule. We then encode both the tree and graph into their latent embeddings zT and zG. To decode the molecule, we first reconstruct junction tree from zT , and then assemble nodes in the tree back to the original molecule.'''
with st.sidebar:
sidebar_con = st.empty()
# sidebar_con.empty()
with sidebar_con.container():
ab_sidebar()
with st.expander(':four_leaf_clover: About the author',expanded=True):
st.markdown('')
st.markdown("
",unsafe_allow_html=True)
with st.expander(':star2: About the model',expanded=True):
st.markdown("Based on Junction Tree Variational Autoencoder for Molecular Graph Generation (JTVAE)",unsafe_allow_html=True)
st.markdown("
Citing
",unsafe_allow_html=True)
st.markdown("Paper: [https://arxiv.org/abs/1802.04364](https://arxiv.org/abs/1802.04364)")
st.code('''@misc{jin2019junction,
title={Junction Tree Variational Autoencoder for Molecular Graph Generation},
author={Wengong Jin and Regina Barzilay and Tommi Jaakkola},
year={2019},
eprint={1802.04364},
archivePrefix={arXiv},
primaryClass={cs.LG}
}''')
st.markdown("
Author
",unsafe_allow_html=True)
st.markdown("Wengong Jin, Regina Barzilay, Tommi Jaakkola",unsafe_allow_html=True)
st.markdown("
Abstract
",unsafe_allow_html=True)
st.markdown(descrip_model)
ab = st.columns([1,10,1])
ab[1].markdown("
",unsafe_allow_html=True)
with st.expander(':snowman: :blue[Instruction]'):
guide = """
Steps to optimize a molucule
1. Select from examples, or manually enter a valid SMILES string of a molecule.
2. Configure the settings to generate a new molecule. The new molecule should have a higher penalized LogP value.
- Learning rate: How 'far' from the molecule that you want to search.
- Similarity cutoff: How 'similar' to the molecule that you want to search.
- Number of iterations: Number of generation trials.
Annotation
SMILES - Simplified molecular-input line-entry system
LogP - The log of the partition coefficient of a solute between octanol and water, at near infinite dilution
SA score - Synthetic Accessibility Score (lower is better)
Cycle score - A number of carbon rings of size larger than 6 (lower is better)
Penalized LogP - Standardized score of LogP - SA score - Cycle scoreSimilarity - Molecular similarity is calculated via Morgan fingerprint of radius 2 with Tanimoto similarity
"""
st.markdown(guide,unsafe_allow_html=True)
with st.sidebar:
sidebar_con = st.empty()
# sidebar_con.empty()
with sidebar_con.container():
set_step(0)
oam_sidebar(0)
# oab_sel_container = st.empty()
if 'checked_single' not in st.session_state:
st.session_state.checked_single = 'NO'
# if 'mode' not in st.session_state:
# st.session_state.mode = 0
if 'single_optimized' not in st.session_state:
st.session_state.single_optimized = False
if 'smiles_checked' not in st.session_state:
st.session_state.smiles_checked = False
if 'compared' not in st.session_state:
st.session_state.compared = False
# with oab_sel_container.container():
sample_mode = {
'-':'',
'Sorafenib':Chem.CanonSmiles('CNC(=O)C1=NC=CC(=C1)OC2=CC=C(C=C2)NC(=O)NC3=CC(=C(C=C3)Cl)C(F)(F)F'),
'Pazopanib':Chem.CanonSmiles('CC1=C(C=C(C=C1)NC2=NC=CC(=N2)N(C)C3=CC4=NN(C(=C4C=C3)C)C)S(=O)(=O)N'),
'Sunitinib':Chem.CanonSmiles('CCN(CC)CCNC(=O)C1=C(NC(=C1C)C=C2C3=C(C=CC(=C3)F)NC2=O)C'),
'Cabozantinib':Chem.CanonSmiles('COC1=CC2=C(C=CN=C2C=C1OC)OC3=CC=C(C=C3)NC(=O)C4(CC4)C(=O)NC5=CC=C(C=C5)F'),
'Axitinib':Chem.CanonSmiles('CNC(=O)C1=CC=CC=C1SC2=CC3=C(C=C2)C(=NN3)C=CC4=CC=CC=N4'),
'Lenvatinib':Chem.CanonSmiles('COC1=CC2=NC=CC(=C2C=C1C(=O)N)OC3=CC(=C(C=C3)NC(=O)NC4CC4)Cl'),
'Regorafenib':Chem.CanonSmiles('CNC(=O)C1=NC=CC(=C1)OC2=CC(=C(C=C2)NC(=O)NC3=CC(=C(C=C3)Cl)C(F)(F)F)F'),
'Vandetanib':Chem.CanonSmiles('CN1CCC(CC1)COC2=C(C=C3C(=C2)N=CN=C3NC4=C(C=C(C=C4)Br)F)OC'),
'Tivozanib': Chem.CanonSmiles('CC1=CC(=NO1)NC(=O)NC2=C(C=C(C=C2)OC3=C4C=C(C(=CC4=NC=C3)OC)OC)Cl')
}
ls_opt = list(sample_mode.keys())
oam_sel_col = st.columns([3,7])
with st.form('sel_smiles'):
mode = oam_sel_col[0].selectbox("Select an example",options=ls_opt,on_change=reset_oam_state)
smiles = oam_sel_col[1].text_input('Enter a SMILES string (max 200 chars):',sample_mode[mode],max_chars=200,
disabled=(mode != '-'))
# if mode == '-':
# st.session_state.smiles = oam_sel_col[1].text_input('Enter a SMILES string (max 200 chars):',max_chars=200,key='opt_0')
# # st.session_state.mode = 0
# elif mode == 'Sorafenib':
# st.session_state.smiles = 'CNC(=O)C1=NC=CC(=C1)OC2=CC=C(C=C2)NC(=O)NC3=CC(=C(C=C3)Cl)C(F)(F)F'
# oam_sel_col[1].text_input('Enter a SMILES string (max 200 chars):','CNC(=O)C1=NC=CC(=C1)OC2=CC=C(C=C2)NC(=O)NC3=CC(=C(C=C3)Cl)C(F)(F)F',max_chars=200,disabled=True,key='opt_1')
# # st.session_state.mode = 1
# elif mode == 'Pazopanib':
# st.session_state.smiles = 'CC1=C(C=C(C=C1)NC2=NC=CC(=N2)N(C)C3=CC4=NN(C(=C4C=C3)C)C)S(=O)(=O)N'
# oam_sel_col[1].text_input('Enter a SMILES string (max 200 chars):','CC1=C(C=C(C=C1)NC2=NC=CC(=N2)N(C)C3=CC4=NN(C(=C4C=C3)C)C)S(=O)(=O)N',max_chars=200,disabled=True,key='opt_2')
# # st.session_state.mode = 2
# elif mode == 'Sunitinib':
# st.session_state.smiles = 'CCN(CC)CCNC(=O)C1=C(NC(=C1C)C=C2C3=C(C=CC(=C3)F)NC2=O)C'
# oam_sel_col[1].text_input('Enter a SMILES string (max 200 chars):','CCN(CC)CCNC(=O)C1=C(NC(=C1C)C=C2C3=C(C=CC(=C3)F)NC2=O)C',max_chars=200,disabled=True,key='opt_3')
# # st.session_state.mode = 3
check_single_butt = st.form_submit_button('Check SMILES',use_container_width=True)
# st.session_state.smiles = st.session_state['opt_'+str(ls_opt.index(st.session_state.mode))]
if check_single_butt:
st.session_state.mode = mode
st.session_state.smiles = smiles
check_single(st.session_state.smiles)
if 'optim_single_butt' not in locals():
optim_single_butt = False
check_single_con = st.empty()
if 'smiles_selected' in st.session_state:
if st.session_state.smiles_selected:
with check_single_con.container():
if 'checked_single' in st.session_state:
if st.session_state.checked_single == 'EnterError':
st.markdown("
",unsafe_allow_html=True)
else:
# st.write('THIS MOLECULE DOES NOT EXIST!')
st.markdown("
THIS MOLECULE DOES NOT EXIST!
",unsafe_allow_html=True)
st.markdown("
Checked using molbloom
",unsafe_allow_html=True)
if st.button('Compare',use_container_width=True):
st.session_state.compared = True
if st.session_state.compared:
compare_single_con.empty()
with compare_single_con.container():
com_col = st.columns(3)
com_col[1].markdown("
"%(new_mol_metrics[met]),unsafe_allow_html=True)
for met,met_name in rule_dict.items():
value_com_col[0].markdown(f"
{met_name}
",unsafe_allow_html=True)
old_passed = old_mol_metrics[met]
new_passed = new_mol_metrics[met]
if met == 'ro5':
value_com_col[1].markdown("
%s
"%('mediumseagreen' if old_passed > 3 else 'tomato',old_passed),unsafe_allow_html=True)
value_com_col[2].markdown("
%s
"%('mediumseagreen' if new_passed > 3 else 'tomato',new_passed),unsafe_allow_html=True)
else:
value_com_col[1].markdown("
%s
"%('mediumseagreen' if old_passed else 'tomato','Passed' if old_passed else 'Failed'),unsafe_allow_html=True)
value_com_col[2].markdown("
%s
"%('mediumseagreen' if new_passed else 'tomato','Passed' if new_passed else 'Failed'),unsafe_allow_html=True)
score_col_old = value_com_col[1].columns(2)
score_col_new = value_com_col[2].columns(2)
for met,met_name in score_dict.items():
value_com_col[0].markdown(f"
{met_name}
",unsafe_allow_html=True)
if met != 'npscore':
score_col_old[0].markdown("
"%('mediumseagreen' if old_passed else 'tomato',old_passed),unsafe_allow_html=True)
# value_com_col[2].markdown("
%s
"%('mediumseagreen' if new_passed else 'tomato',new_passed),unsafe_allow_html=True)
for met,met_name in filter_dict.items():
# value_com_col[0].markdown(f"
",unsafe_allow_html=True)
with st.expander(':snowman: :blue[Instruction]'):
guide = """
Steps to optimize a molucule
1. Upload a text file with SMILES string on each line.
2. Check the SMILES strings to make sure that they are valid and pass MOSES filters.
3. Select scores to calculate (penalized LogP included). Keep passed SMILES and calculate selected scores.
4. Configure the settings to generate new molecules. The new molecules should have higher penalized LogP values.
- Learning rate: How 'far' from each molecule that you want to search
- Similarity cutoff: How 'similar' to each molecule that you want to search
- Number of iterations: Number of generation trials per molecule
5. (Optional) You can download the dataframe at any steps as *.csv file.
Annotation
SMILES - Simplified molecular-input line-entry system
LogP - The log of the partition coefficient of a solute between octanol and water, at near infinite dilution
SA score - Synthetic Accessibility Score (lower is better)
Cycle score - A number of carbon rings of size larger than 6 (lower is better)
Penalized LogP - Standardized score of LogP - SA score - Cycle scoreSimilarity - Molecular similarity is calculated via Morgan fingerprint of radius 2 with Tanimoto similarity
"""
st.markdown(guide,unsafe_allow_html=True)
with st.sidebar:
sidebar_con = st.empty()
# sidebar_con.empty()
with sidebar_con.container():
set_step(0)
oab_sidebar(0)
oab_upl_container = st.empty()
if 'smiles_upload_change' not in st.session_state:
st.session_state.smiles_upload_change = False
if 'checked_batch' not in st.session_state:
st.session_state.checked_batch = False
if 'batch_left_checked' not in st.session_state:
st.session_state.batch_left_checked = False
if 'scores_calculated' not in st.session_state:
st.session_state.scores_calculated = False
if 'batch_optimized' not in st.session_state:
st.session_state.batch_optimized = False
with oab_upl_container.container():
st.session_state['smiles_file'] = st.file_uploader("Upload a text file with SMILES on each line :sparkles:",on_change=reset_oab_state)
if 'check_batch_butt' not in locals():
check_batch_butt = False
if st.session_state['smiles_file'] is not None:
if st.session_state.smiles_upload_change:
smiles_list = io.StringIO(st.session_state.smiles_file.getvalue().decode("utf-8"))
smiles_list = list(smiles_list.getvalue().rstrip().split('\n'))
st.markdown('Number of SMILES: '+str(len(smiles_list)))
if len(smiles_list) == 1:
st.markdown("
Please use Optimize a molecule tab.
",unsafe_allow_html=True)
with sidebar_con.container():
set_step(0)
oab_sidebar(0)
else:
st.session_state['df'] = pd.DataFrame({'SMILES':smiles_list})
st.dataframe(st.session_state['df'],use_container_width=True)
check_batch_butt = st.button('Check SMILES')
else:
# if not st.session_state.checked_batch:
if st.session_state['smiles_file'] is not None:
st.dataframe(st.session_state['df'],use_container_width=True)
# st.button('Check SMILES',on_click=check_batch,args=[smiles_list],key='check_batch_butt')
if check_batch_butt:
if st.session_state.smiles_upload_change:
with sidebar_con.container():
set_step(1)
oab_sidebar(1)
check_batch(list(st.session_state['df'].SMILES))
st.session_state.smiles_upload_change = False
if 'calc_batch_butt' not in locals():
calc_batch_butt = False
check_batch_con = st.empty()
calc_batch_con = st.empty()
if st.session_state.checked_batch:
with check_batch_con.container():
passed_num = st.session_state['df'][st.session_state['df'].checked != 'invalid'].shape[0]
st.markdown('Number of passed SMILES: '+str(passed_num))
st.dataframe(st.session_state['df'].style.applymap(highlight_result, subset=pd.IndexSlice[:, ['checked']]),use_container_width=True)
if passed_num == 0:
st.markdown("
The uploaded file contains no suitable SMILES string.
",unsafe_allow_html=True)
st.session_state.batch_left_checked = False
with sidebar_con.container():
set_step(0)
oab_sidebar(0)
else:
st.session_state.batch_left_checked = True
df = st.session_state['df']
download_df(df,0)
choose_score_con = st.empty()
if st.session_state.batch_left_checked:
with sidebar_con.container():
set_step(2)
oab_sidebar(2)
with choose_score_con.container():
with st.form("Choose score to calculate"):
st.markdown("
Choose score(s) to calculate
",unsafe_allow_html=True)
st.caption('Penalized LogP is always calculated.')
st.checkbox('LogP',key='logp_cal')
st.checkbox('SA score',key='sa_cal')
st.checkbox('Cycle score',key='cycle_cal')
calc_batch_butt = st.form_submit_button("Keep passed SMILES and calculate scores")
else:
choose_score_con.empty()
else:
check_batch_con.empty()
if 'optim_batch_butt' not in locals():
optim_batch_butt = False
# if 'calc_batch_butt' in st.session_state:
if calc_batch_butt and st.session_state.batch_left_checked:
# if not st.session_state.scores_calculated:
smiles_list = list(st.session_state['df'][st.session_state['df'].checked != 'invalid'].checked)
st.session_state.score_df = calc_scores(smiles_list)
st.session_state.batch_optimized = False
if st.session_state.scores_calculated:
calc_batch_con.empty()
with calc_batch_con.container():
st.dataframe(st.session_state.score_df,use_container_width=True)
score_df = st.session_state.score_df
download_df(score_df,1)
with sidebar_con.container():
set_step(3)
oab_sidebar(3)
with st.form(":gear: Settings"):
st.slider('Choose learning rate: ',0.0,5.0,0.4,key='lr_b')
st.slider('Choose similarity cutoff: ',0.0,1.0,0.4,key='sim_cutoff_b')
st.slider('Choose number of iterations: ',1,100,80,key='n_iter_b')
optim_batch_butt = st.form_submit_button("Optimize")
else:
calc_batch_con.empty()
optim_batch_con = st.empty()
ani_con = st.empty()
if optim_batch_butt and st.session_state.scores_calculated:
optim_batch_con.empty()
with sidebar_con.container():
set_step(4)
oab_sidebar(4)
with ani_con.container():
st.markdown('Operation in progress. Please wait...')
gen_results = []
render_animation()
st.markdown('Generating new SMILES string(s)...')
model = load_model()
for canon_smiles in stqdm(list(st.session_state.score_df.SMILES)):
gen_results.append(optim_single(canon_smiles,model,st.session_state.lr_b,st.session_state.sim_cutoff_b,st.session_state.n_iter_b))
st.markdown('Checking generated SMILES string(s) ...')
st.session_state.new_score_df = calc_scores_new(gen_results)
ani_con.empty()
if st.session_state.batch_optimized:
with sidebar_con.container():
set_step(5)
oab_sidebar(5)
with optim_batch_con.container():
new_score_df = st.session_state.new_score_df
# new_score_df.style.applymap(highlight_result, subset=pd.IndexSlice[:, ['new_smiles']])
st.markdown("
RESULTS
",unsafe_allow_html=True)
st.dataframe(new_score_df.style.applymap(highlight_result, subset=pd.IndexSlice[:, ['new_smiles']]),use_container_width=True)
download_df(new_score_df,3)
else:
optim_batch_con.empty()
def process_check_single(smiles):
mol = Chem.MolFromSmiles(smiles)
if (mol is not None) and (mol_passes_filters_custom(mol) == 'YES'):
return Chem.MolToSmiles(mol)
else:
return 'invalid'
def check_batch(smiles_list):
check = []
# check = Parallel(n_jobs=-1,backend='loky')(
# delayed(process_check_single)(smi) for smi in stqdm(smiles_list)
# )
for smi in stqdm(smiles_list):
mol = Chem.MolFromSmiles(smi)
if (mol is not None) and (mol_passes_filters_custom(mol) == 'YES'):
check.append(Chem.MolToSmiles(mol))
else:
check.append('invalid')
st.session_state['df'] = pd.concat([st.session_state['df'],pd.DataFrame({'checked':check})],axis=1)
st.session_state.checked_batch = True
# return check
def calc_scores(smiles_list):
score_df = pd.concat([st.session_state.df[st.session_state.df.checked != 'invalid'].SMILES,pd.DataFrame({'Checked_SMILES':smiles_list})],axis=1)
scores =[]
# scores = Parallel(n_jobs=-1,backend='loky')(
# delayed(penalized_logp_standard)(Chem.MolFromSmiles(smi)) for smi in stqdm(smiles_list)
# )
for smi in stqdm(smiles_list):
logp,sa,cycle,pen_logp = penalized_logp_standard(Chem.MolFromSmiles(smi))
scores+=[(logp,sa,cycle,pen_logp)]
s_df = pd.DataFrame(scores,columns=st.session_state.sc_name)
for n, checked in zip(st.session_state.sc_name,[st.session_state.logp_cal,st.session_state.sa_cal,st.session_state.cycle_cal,True]):
if checked:
score_df = pd.concat([score_df,s_df[n]],axis=1)
st.session_state.scores_calculated = True
return score_df
def process_calc_new_score(new_smiles,sim):
if new_smiles is None:
return ('invalid',-100.0,-100.0,-100.0,-100.0,-100.0)
else:
new_mol = Chem.MolFromSmiles(new_smiles)
if new_mol is None:
return ('invalid',-100.0,-100.0,-100.0,-100.0,-100.0)
else:
logp,sa,cycle,pen_logp = penalized_logp_standard(new_mol)
return (new_smiles,sim,logp,sa,cycle,pen_logp)
def calc_scores_new(result):
new_scores =[]
# new_scores = Parallel(n_jobs=-1,backend='loky')(
# delayed(process_calc_new_score)(new_smiles,sim) for new_smiles,sim in stqdm(result)
# )
for new_smiles,sim in stqdm(result):
if new_smiles is None:
new_scores+=[('invalid',-100.0,-100.0,-100.0,-100.0,-100.0)]
else:
new_mol = Chem.MolFromSmiles(new_smiles)
if new_mol is None:
new_scores+=[('invalid',-100.0,-100.0,-100.0,-100.0,-100.0)]
else:
logp,sa,cycle,pen_logp = penalized_logp_standard(new_mol)
new_scores+=[(new_smiles,sim,logp,sa,cycle,pen_logp)]
new_col = ['new_smiles','sim']+st.session_state.new_sc_name
s_df = pd.DataFrame(new_scores,columns=new_col)
new_score_df = st.session_state.score_df
for n, checked in zip(new_col,[True, True,st.session_state.logp_cal,st.session_state.sa_cal,st.session_state.cycle_cal,True]):
if checked:
new_score_df = pd.concat([new_score_df,s_df[n]],axis=1)
st.session_state.batch_optimized = True
return new_score_df
def highlight_result(value):
if value == 'invalid': color = 'tomato'
else: color = 'mediumseagreen'
return 'color: %s' % color
@st.cache_data(experimental_allow_widgets=True)
def download_df(df,id):
with st.expander(':arrow_down: Download this dataframe'):
st.markdown("
Select column(s) to save:
",unsafe_allow_html=True)
for col in df.columns:
st.checkbox(col,key=str(id)+'_col_'+str(col),value=True)
st.text_input('File name (.csv):','dataframe',key=str(id)+'_file_name')
ste.download_button('Download',df_to_file(df[[col for col in df.columns if st.session_state[str(id)+'_col_'+str(col)]]]),st.session_state[str(id)+'_file_name']+'.csv')
def reset_oam_state():
st.session_state.smiles_selected = False
st.session_state.checked_single = 'NO'
st.session_state.smiles_checked = False
st.session_state.single_optimized = False
st.session_state.compared = False
set_step(0)
def reset_oab_state():
st.session_state.smiles_upload_change = True
st.session_state.smiles_uploaded = False
st.session_state.checked_batch = False
st.session_state.batch_left_checked = False
st.session_state.scores_calculated = False
st.session_state.batch_optimized = False
set_step(0)
def rerun():
st.experimental_rerun()
def render_view():
# render_sidebar(st.session_state.current_view,st.session_state.current_step)
form_header()
form_body()
render_view()
).decode()})
"+
"
".format(
img_to_bytes(img_path), max_width
)
return img_html
_mcf = pd.read_csv('./mcf.csv')
_pains = pd.read_csv('./wehi_pains.csv',
names=['smarts', 'names'])
_mcf_filters = [Chem.MolFromSmarts(x) for x in
_mcf['smarts'].values]
_pains_filters = [Chem.MolFromSmarts(x) for x in
_pains['smarts'].values]
def mol_passes_filters_custom(mol,
allowed=None,
isomericSmiles=False):
"""
Checks if mol
* passes MCF and PAINS filters,
* has only allowed atoms
* is not charged
"""
allowed = allowed or {'C', 'N', 'S', 'O', 'F', 'Cl', 'Br', 'H'}
if mol is None:
return 'NoMol'
ring_info = mol.GetRingInfo()
if ring_info.NumRings() != 0 and any(
len(x) >= 8 for x in ring_info.AtomRings()
):
return 'ManyRings'
h_mol = Chem.AddHs(mol)
if any(atom.GetFormalCharge() != 0 for atom in mol.GetAtoms()):
return 'Charged'
if any(atom.GetSymbol() not in allowed for atom in mol.GetAtoms()):
return 'AtomNotAllowed'
if any(h_mol.HasSubstructMatch(smarts) for smarts in _mcf_filters):
return 'MCF'
if any(h_mol.HasSubstructMatch(smarts) for smarts in _pains_filters):
return 'PAINS'
smiles = Chem.MolToSmiles(mol, isomericSmiles=isomericSmiles)
if smiles is None or len(smiles) == 0:
return 'Isomeric'
if Chem.MolFromSmiles(smiles) is None:
return 'Isomeric'
if not check_vocab(Chem.MolToSmiles(mol)):
return 'NoVocab'
return 'YES'
def penalized_logp_standard(mol):
logP_mean = 2.4399606244103639873799239
logP_std = 0.9293197802518905481505840
SA_mean = -2.4485512208785431553792478
SA_std = 0.4603110476923852334429910
cycle_mean = -0.0307270378623088931402396
cycle_std = 0.2163675785228087178335699
log_p = Descriptors.MolLogP(mol)
SA = -sascorer.calculateScore(mol)
# cycle score
cycle_list = nx.cycle_basis(nx.Graph(Chem.rdmolops.GetAdjacencyMatrix(mol)))
if len(cycle_list) == 0:
cycle_length = 0
else:
cycle_length = max([len(j) for j in cycle_list])
if cycle_length <= 6:
cycle_length = 0
else:
cycle_length = cycle_length - 6
cycle_score = -cycle_length
# print(logP_mean)
standardized_log_p = (log_p - logP_mean) / logP_std
standardized_SA = (SA - SA_mean) / SA_std
standardized_cycle = (cycle_score - cycle_mean) / cycle_std
return log_p,SA,cycle_score,standardized_log_p + standardized_SA + standardized_cycle
def df_to_file(df):
s_buff.seek(0)
df.to_csv(s_buff)
return s_buff.getvalue().encode()
# def download_df(df,id):
# with st.expander(':arrow_down: Download this dataframe'):
# st.markdown("
"+
"
"+
"
"+
"