Spaces:
Runtime error
Runtime error
File size: 33,777 Bytes
356b18e 66ab79e 356b18e 66ab79e 50199cc 356b18e 66ab79e 356b18e 66ab79e 356b18e 66ab79e 356b18e 66ab79e 356b18e 66ab79e 356b18e 66ab79e 356b18e 66ab79e 356b18e 66ab79e 356b18e 66ab79e 356b18e 66ab79e 356b18e 66ab79e 356b18e 9f8cc36 356b18e 9f8cc36 22bda93 9f8cc36 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 |
import numpy as np
import pickle
import pandas as pd
# import requests
# from selenium import webdriver
import matplotlib.pyplot as plt
#Simple assignment
# from selenium.webdriver import Firefox
# from selenium.webdriver.common.keys import Keys
# from selenium.common.exceptions import NoSuchElementException
# import requests
import os
import seaborn as sns
from collections import Counter
import plotly.express as px
import streamlit as st
### Scrap the cosmic id information
# ### FRAMEWORKS NEEDED
# def scrap():
# #### Setting options to the driver
# options = webdriver.FirefoxOptions()
# options.add_argument('--headless')
# options.add_argument('--no-sandbox')
# options.add_argument('--disable-dev-shm-usage')
# options.capabilities
# ### Setting options of webdriver
# # a) Setting the chromedriver
# browser = Firefox(options=options,executable_path=r"C:\Users\Pablo\OneDrive\Documents\Documentos\Escuela Politécnica Superior Leganés\4 AÑO\ASIGNATURAS\1 CUATRI\WEB ANALYTICS\PART 2\Milestone3\geckodriver.exe")
# ### Functions and execution to run the scrapping
# def getinfofromtable(oddrows:list,score:float,headertable)->list:
# rows = []
# for row in oddrows:
# cols = []
# for (i,col) in enumerate(row.find_elements_by_css_selector("td")):
# if i==headertable.index( 'Primary Tissue') or i==headertable.index('Primary Histology') or i==headertable.index('Zygosity'):
# cols.append(col.text)
# cols.append(score)
# rows.append(cols)
# return rows
# def getinfocosmic(mutationid):
# import time
# search = browser.find_element_by_id('search-field')
# search = search.find_element_by_class_name("text_def")
# search.send_keys(mutationid)
# search.send_keys(Keys.RETURN)
# time.sleep(5)
# try:
# container = browser.find_element_by_id("section-list")
# except NoSuchElementException:
# return []
# try:
# subq1 = container.text[container.text.find("score")+len("score"):]
# score = float(subq1[:subq1.find(")")].strip())
# except ValueError:
# score = 0
# section = browser.find_element_by_id("DataTables_Table_0")
# headertable = [header.text for header in section.find_element_by_tag_name("thead").find_elements_by_tag_name("th")]
# oddrows = section.find_elements_by_class_name("odd")
# evenrows = section.find_elements_by_class_name("even")
# l1 = getinfofromtable(oddrows,score,headertable)
# l1.extend(getinfofromtable(evenrows,score,headertable))
# # browser.close()
# return l1
# ## Looking for cosmic id info
# cosl = []
# browser.get("https://cancer.sanger.ac.uk/cosmic")
# for cos in cosmicinfo.reset_index()["COSMIC_ID"].iloc[20:]:
# if cos.find(",")!=-1:
# cos = cos.split(",")[0]
# cosl.append(getinfocosmic(cos))
# browser.get("https://cancer.sanger.ac.uk/cosmic")
### Pieplots
def pieplot(merging,id=0):
genecount = merging.groupby(by=["gene_name","UV_exposure_tissue","sampleID"]).count().reset_index()
if id==0:
gtype = genecount[genecount.UV_exposure_tissue=="Intermittently-photoexposed"]
if id ==1 :
gtype = genecount[genecount.UV_exposure_tissue=="Chronically-photoexposed"]
else:
gtype = genecount
gtype = gtype.groupby("gene_name").count()["sampleID"].reset_index()
gtype.sort_values(by="sampleID",ascending=False,inplace=True)
#define Seaborn color palette to use
colors = sns.color_palette('pastel')[0:len(gtype)]
#create pie chart
# plt.suptitle("Gene Occuring for different genes")
plt.pie(gtype.sampleID, labels =gtype.gene_name, colors = colors, autopct='%.0f%%',radius=2,textprops={"fontsize":9})
plt.show()
### Depending on what result you want you return one or another
def filterp4(dfgenes,id=0):
if id==0 or id==1:
if id==0:
chexposed= dfgenes[dfgenes.UV_exposure_tissue=="Intermittently-photoexposed"].sort_values(by=["mean_mut"],ascending=False)
if id==1:
chexposed= dfgenes[dfgenes.UV_exposure_tissue=="Chronically-photoexposed"].sort_values(by=["mean_mut"],ascending=False)
return px.bar(chexposed,x="gene_name",y="mean_mut",error_y="std")
if id==2:
return px.bar(dfgenes,x="gene_name",y="mean_mut",color="UV_exposure_tissue",barmode='group',error_y="std")
### Read scrapping done with cosmic ids
def read_scrap()->list:
with open('my_pickle_file.pickle', 'rb') as f :
cosbase = pickle.load(f)
return cosbase
### GendfClean
def gendfclean(cosbase,cid)->pd.DataFrame:
dfd = {"tissue": None , "histology": None,"zygosity": None, "score": None }
for i,key in enumerate(list(dfd.keys())):
dfd[key] = list(map(lambda x : np.array(x)[:,i].tolist() if x!=[] else [] ,cosbase))
dfd["cosmic_id"] = cid.tolist()
cosmicdb = pd.DataFrame(dfd)
cosmicdb = cosmicdb[(cosmicdb['tissue'].map(lambda d: len(d)) > 0) & (cosmicdb['histology'].map(lambda d: len(d)) > 0) & (cosmicdb['zygosity'].map(lambda d: len(d)) > 0) & (cosmicdb['score'].map(lambda d: len(d)) > 0) ]
cosmicdb["score"] = cosmicdb.score.apply(lambda x: float(x[0]))
return cosmicdb
### Look for stats of a gene
def inputgene(lookforgene,merging,id =0)->dict:
### id = 0--> Intermittently exposed
### id = 1--> Continuously exposed
genecount = merging.groupby(by=["gene_name","UV_exposure_tissue","sampleID"]).count().reset_index()
tgene = genecount[genecount.gene_name==lookforgene]
if id==0:
ph_gene = tgene[tgene.UV_exposure_tissue=='Intermittently-photoexposed']
else:
ph_gene = tgene[tgene.UV_exposure_tissue=="Chronically-photoexposed"]
### Statistiacs about gene|samples
stats = ph_gene.chr.describe()
dc = dict(stats)
dc["gene_name"] = lookforgene
if id==0:
dc["UV_exposure_tissue"] = 'Intermittently-photoexposed'
else:
dc["UV_exposure_tissue"] = 'Chronically-photoexposed'
return dc
### Look for stats of all genes
def gene_exposed(merging,id=0):
return pd.DataFrame(list(map(lambda gene: inputgene(gene,merging,id),merging.gene_name.unique())))
### Merge stats for continuous and intermittently exposed
def mergecontintinfo(merging):
### Continuously Exposed
cont_exposed_info = gene_exposed(merging,1)
### Intermittently Exposed
int_exposed_info = gene_exposed(merging,0)
return pd.concat([cont_exposed_info,int_exposed_info],axis=0)
#### Common tissues, zygosities and histologies
def explodecommon(bd,N,col):
return Counter(bd[col].apply(lambda x: list(x.keys())).explode()).most_common(N)
def pdcommon(db,col,uv:str)->pd.DataFrame:
df = pd.DataFrame(db).rename(columns={0:col,1:"Times_{}".format(col)})
df["UV_exposure_tissue"] = uv
return df
def get_N_common(df,col,N=10)->pd.DataFrame:
cosm = df.copy(True)
cosm[col] = cosm[col].apply(lambda x: Counter(x))
intcosm = cosm[cosm.UV_exposure_tissue=="Intermittently-photoexposed"]
contcosm = cosm[cosm.UV_exposure_tissue=="Chronically-photoexposed"]
infotissues = explodecommon(cosm,N,col)
inttissues = explodecommon(intcosm,N,col)
contissues = explodecommon(contcosm,N,col)
df1 = pdcommon(infotissues,col,"Total")
df2 = pdcommon(inttissues,col,"Intermittently-photoexposed")
df3 = pdcommon(contissues,col,"Chronically-photoexposed")
return pd.concat([df1,df2,df3],axis=0)
### Deatiled information of mutation type
def mut_type(x):
if x.mut_type=="Indel":
if len(x.ref)>len(x.mut):
return "Del"
elif len(x.mut)>len(x.ref):
return "In"
# if len(x.ref)>1 and len(x.mut)>1:
return x.ref+">"+x.mut
return x.mut_type
def distribution_gene(df,hue):
plot4 = df.groupby([hue,"mut_type_cus"]).count().reset_index().iloc[:,:3]
plot4 = plot4.rename(columns={"sampleID":"n_mut"})
plot4 = plot4.sort_values(by="mut_type_cus",ascending=True)
fig = px.bar(plot4,x="mut_type_cus",y="n_mut",color=hue,barmode="group")
return fig
directory = os.path.abspath("")
# from EDA_IMDb_functions import *
st.set_page_config(layout="wide")
st.set_option('deprecation.showPyplotGlobalUse', False)
dw,col1,wl = st.columns((1,0.5,1))
col1.image('img/descarga.jfif')
st.markdown("<h1 style='text-align:center;'>Somatic Mutations Analysis in skin</h1>",unsafe_allow_html=True)
st.sidebar.markdown("<h2 style='text-align:center;'>Index</h2>",unsafe_allow_html=True)
menu = st.sidebar.radio(
"",
("1. Intro", "2. Analysis of somatic mutations" ,'3. Sample Analysis (IGV)'),
)
# Pone el radio-button en horizontal. Afecta a todos los radio button de una página.
# Por eso está puesto en este que es general a todo
# st.write('<style>div.row-widget.stRadio > div{flex-direction:row;}</style>', unsafe_allow_html=True)
st.sidebar.markdown('---')
st.sidebar.markdown("<h2 style='text-align:center;'>Authors</h2>",unsafe_allow_html=True)
st.sidebar.markdown("<p style='text-align:center;'>Claudio Sotillos Peceroso</p>",unsafe_allow_html=True)
st.sidebar.markdown("<p style='text-align:center;'>Pablo Reyes Martin</p>",unsafe_allow_html=True)
@st.cache(allow_output_mutation=True)
def read_csv():
return pd.ExcelFile('Study Results.xlsx')
### Merge two dataframes
def definemerging(df1,df2):
merging = df1.merge(df2,on="sampleID",how="inner")
return merging
#### Reading the data
df = read_csv()
df1 = pd.read_excel(df, 'Dataset_S1').copy(deep=True)
df2 = pd.read_excel(df, 'Dataset_S2').copy(deep=True)
merging = None
### functions indexed
def wrt(text,tag="h1",align="center"):
return st.markdown(f"""
<{tag} style='text-align:{align};'>{text}</{tag}>
""",unsafe_allow_html=True)
def writetxt(text,tag="h1",align="center",container=st):
return container.markdown(f"""
<{tag} style='text-align:{align};'>{text}</{tag}>
""",unsafe_allow_html=True)
def space(tag="h1"):
return wrt("\n",tag=tag)
### 1. Introduciton
def set_home():
wrt("1. Introduction to the problem",tag="h2")
par1 = """
In this project, we have focus our study on the research paper as well as on
getting some conclusion by ourselves. Therefore there will be some contrast we will do in relation to the paper
as well as other plots not related with the paper. Also, we will perform the analysis of two samples that are
from quite opposite individuals."""
### 1.1
wrt(par1,tag="p style='font-size:21px;'",align="justify")
tit1 = """
1.1 Cancers arise as a result of somatic mutations
"""
wrt(tit1,tag="h4",align="left")
___,col1,_,col2,___ = st.columns((0.4,1,0.7,1,0.4))
cap1 = "Fig 1 : Types of Genomic Alterations"
col1.image("img/base.png",caption=cap1,width=500,use_column_width=True)
cap2 = "Fig 2 : Skin Risk Factors"
col2.image("img/risk_factors.png",caption=cap2,width=500,use_column_width=True)
with col2:
col2_par = """
The proportion of somatic mutations in normal cells is quite similar to the mutations of tumours in the same tissue cell.
Only a small fraction are the ones which provokes cancer.
"""
wrt(col2_par,tag="p style='font-size:20px;'",align="justify")
space()
### 1.2
tit1 = """
1.2 State of the art of the research
"""
wrt(tit1,tag="h4",align="left")
___,col1,_,col2,___ = st.columns((0.4,1.4,0.3,0.6,0.4))
cap1 = "Fig 1 : Types of Genomic Alterations"
col1.image("img/comboskin.JPG",caption="Fig 3 : Cutaneous Sensitivity",width=600,use_column_width=True)
# cap2 = "Fig 2 : Skin Risk Factors"
col2.image("img/s_exposure.JPG",width=600,use_column_width=True)
# col1,middle,col2 = st.columns((0.2,3,0.2))
with col2:
col2_par = """
- About 25,50% of the normal skins acquires one driver mutation.
- Chronic sun exposure tends to proliferate keratinocytes (cutaneous squamous cell carcinoma)
- Melanomas appears more in sporadic skin sun exposure (attributable to intermittent pattern of sun exposure with recreational activities)
"""
st.markdown(col2_par)
# wrt(col2_par,tag="p",align="center")
space()
par2= """
The research studies which are the main factors that causes somatic mutations in skin. The research analyzes 46 genes per sample.
Skin samples were collected from different body areas, classified according to the pattern of sunlight exposure as
chronically-photoexposed (n = 44) and intermittently photoexposed (n = 79).
"""
wrt("1.3 Skin samples analysis",tag="h4",align="left")
wrt(par2,tag="p style='font-size:21px;'",align="justify")
import plotly.graph_objects as go
set = ["Chronically Sun Exposure","Inttermittently Sun Exposure"]
n = [44,79]
fig = go.Figure(data=[go.Pie(labels=set, values=n,textinfo=f'label+percent',
insidetextorientation='radial',hole=.3)])
fig.update_layout(title_text ="Samples of the dataset depending on the sun exposure",annotations=[dict(text="46 genes per sample",font_size=20, showarrow=False)])
__,col1,__,col2,__ = st.columns((0.2,1,0.5,0.5,0.2))
col1.plotly_chart(fig,use_container_width=True)
val = round(df1.groupby("sampleID").count()["chr"].mean(),2)
pnt = round(val/46,2)
col2.header("\n\n\n")
col2.header("\n\n\n")
col2.header("\n\n\n")
col2.metric("Average of mutations per sample",val)
col2.metric("Average Percentage of genes that mutate per sample",pnt)
#### 2. Plots Dataframe visualization Conclusion
def set_chintp():
global merging
wrt("2. Analysis of somatic mutations",tag="h3")
par1 = """
We are going to discuss some plots related to the paper. Moreover, we are going to explore mutations of genes that occured within our samples as well as go beyond our data base, exploring
mutations that have cosmic ids so that we can show more detailed information about the mutation. We often face samples that are intermittently photoexposed to sunlight against the ones which are continuously exposed
to contrast two distinct populations.
"""
wrt(par1,tag="p style='font-size:21px;",align="justify")
wrt("2.1 Description of the dataset",tag="h4",align="left")
par1 = """
They focused on sequencing 123 samples of healthy skin (from different areas,permanently or intermittently photo-exposed) of cancer-free
individuals. It was taken just one skin sample per individual.
From each of these samples, a deep sequencing of 46 genes (which are implicated in skin cancer) is carried out.
This means that they had to analyze 5658 genes. Out of these amount of genes they found 5214 somatic mutations, which are the ones
which we have in our dataset.
"""
wrt(par1,tag="p style='font-size:21px;",align="justify")
st.image("img/dataset.png",width=600,use_column_width=True)
st.image("img/skin.png",width=600,use_column_width=True)
wrt("2.2 Average and Standard deviation of mutations",tag="h4",align="left")
par3 ="""
The research tells that if the skin were exposed more to the skin, it will be more likely to suffer somatic mutations.
The first plot represent the average and sd of mutations that have samples which are intermittently and continuously exposed to sunlight.
"""
wrt(par3,tag="p style='font-size:21px;",align="justify")
#### CODE PLOT 1
merging = definemerging(df1,df2)
plot = merging.groupby("sampleID").agg({"gene_name":"count"}).merge(df2[["sampleID","UV_exposure_tissue"]],on="sampleID",how="inner")
plot = plot.rename(columns={"gene_name":"n_of_mutations"})
fig = plt.figure(figsize=(5,3))
sns.barplot(data=plot,x="UV_exposure_tissue",y="n_of_mutations")
plt.suptitle("Average and standard deviation of mutations per type tissue photoexposed")
st.pyplot(fig,clear_figure=True)
wrt("\n")
### Plot 3
# """Number of samples that has a mutation at least one time per gene"""
wrt("2.3 Number of samples per gene captured at least one time",tag="h4",align="left")
par = """
This takes the number of times that a gene appears at least once time in the samples intermittently and continuously exposed.
The first plots is used for the samples that are intermittently exposed and the second plot is used for the ones which are continously exposed
"""
print( merging.UV_exposure_tissue.unique())
wrt(par,tag="p style='font-size:21px;",align="justify")
wrt("Select the exposure tissue you want",tag="h6",align="left")
box = st.selectbox("", merging.UV_exposure_tissue.unique().tolist()+["Total"],key="key1")
if box==merging.UV_exposure_tissue.unique()[0]:
wrt("Gene proportion (Intermittently Exposed)",tag="h5",align="center")
pieplot(merging,0)
st.pyplot()
if box==merging.UV_exposure_tissue.unique()[1]:
wrt("Gene proportion (Continuously Exposed)",tag="h5",align="center")
pieplot(merging,1)
st.pyplot()
if box=="Total":
wrt("Gene proportion (Total)",tag="h5",align="center")
pieplot(merging,2)
st.pyplot()
### Plot 4
dfgenes = mergecontintinfo(merging)
dfgenes["mean_mut"] = dfgenes["mean"]
wrt("2.4 Average and standard deviation of mutations per gene",tag="h4",align="left")
par = """
In this graph we will visualize the average and standard deviation of mutation per gene
"""
wrt(par,tag="p style='font-size:21px;",align="justify")
wrt("Select the exposure tissue you want",tag="h6",align="left")
box2 = st.selectbox("", merging.UV_exposure_tissue.unique().tolist()+["Total"],key="key2")
if box2==merging.UV_exposure_tissue.unique()[0]:
fig = filterp4(dfgenes)
if box2==merging.UV_exposure_tissue.unique()[1]:
fig = filterp4(dfgenes,1)
if box2=="Total":
fig = filterp4(dfgenes,2)
st.plotly_chart(fig,use_container_width=True)
#### SECTION 2.2
wrt("2.5 Number of mutations per gene and skin phototype",tag="h4",align="left")
par ="""
These visualizations reflectas the average of mutations that are per age of samples and is divided per skin_phototype
"""
wrt(par,tag="p style='font-size:21px;",align="justify")
#### CODE PLOT 2
col1,col2 = st.columns((1,1))
plot2= df2.merge(merging.groupby("sampleID").agg({"gene_name":"count"}).reset_index(),on="sampleID",how="inner")
plot2 =plot2.rename(columns={"gene_name":"n_mutations"})
plot2.n_mutations = plot2.n_mutations.astype(int)
### plot 2.1
sns.set()
plt.figure()
sns.lmplot(data=plot2.sort_values(by="skin_phototype"),x="age",y="n_mutations",col="skin_phototype",lowess=True,col_wrap=2)
col1.pyplot()
### plot 2.2
sns.set()
plt.figure()
sns.lmplot(data=plot2,x="age",y="n_mutations",hue="skin_phototype",lowess=True)
plt.title("Regression of number of mutations per skin phototype")
col2.pyplot()
writetxt("""Fig 4:Number of mutations per gene and skin phototype""",tag="h6",container=col2 )
wrt("2.6 Number of mutations per exposure tissue condition and sun damage tissue",tag="h4",align="left")
par ="""
We take the average of mutations that we have per type of sample (exposure_tissue type) and if it presents sun damage tissue a priori or not
"""
wrt(par,tag="p style='font-size:21px;",align="justify")
fig = plt.figure(figsize=(5,1))
p = sns.displot(df2, x="UV_exposure_tissue", hue="sun_damage_tissue", multiple="dodge",height=3,aspect=4)
p.fig.set_dpi(100)
st.pyplot(clear_figure=True)
wrt("2.7 Influence of sun damage tissue",tag="h4",align="left")
par ="""
We can see that the sun damage tissue influence generally on the number of mutations in the gene.
"""
wrt(par,tag="p style='font-size:21px;",align="justify")
fig = plt.figure(figsize=(5,1))
p=sns.displot(df2, x="age", hue="sun_damage_tissue", multiple="dodge",height=3,aspect=4).set(title='Age VS Sun Damage')
p.fig.set_dpi(100)
st.pyplot(clear_figure=True)
#### Age vs uv photexposure
wrt("2.8 Number of mutations per year and UV_exposure tissue",tag="h4",align="left")
par =""" We can distinguish between two sample the average of mutations that are got per year"""
wrt(par,tag="p style='font-size:21px;",align="justify")
fig = plt.figure(figsize=(5,2))
# sns.set(rc={'figure.figsize':(3,1),"figure.height":2})
ax = sns.kdeplot(data=df2, x="age", hue="UV_exposure_tissue")
plt.setp(ax.get_legend().get_texts(), fontsize='5') # for legend text
plt.setp(ax.get_legend().get_title(), fontsize='8') # for legend title
st.pyplot()
### Mut type cus plot
wrt("2.9 How is the mutation in comparison with the reference base ?",tag="h4",align="left")
par ="""
We want to know between all the mutations occured (with and without cosmic id) what are the most common mutation type occured
"""
wrt(par,tag="p style='font-size:21px;",align="justify")
merging["mut_type_cus"] = merging.apply(mut_type,axis=1)
hue = st.selectbox("Select label for you want to color: ",['sex', 'UV_exposure_tissue', 'sun_damage_tissue',
'sun_history', 'skin_phototype'],index=1)
fig = distribution_gene(merging,hue)
# plot4 = merging.groupby(["UV_exposure_tissue","mut_type_cus"]).count().reset_index().iloc[:,:3]
# plot4 = plot4.rename(columns={"sampleID":"n_mut"})
# plot4 = plot4.sort_values(by="mut_type_cus",ascending=True)
# fig = px.bar(plot4,x="mut_type_cus",y="n_mut",color="UV_exposure_tissue",barmode="group")
st.plotly_chart(fig,use_container_width=True)
### Muations cosmic and non cosmic id
wrt("2.10 Where does mutations usually occur ?",tag="h4",align="left")
par = "In this section we visualize where the mutations usually occurs as well as tumours that may be caused these mutations. Note that we can get this information just for all the samples that contains cosmic id."
wrt(par,tag="p style='font-size:21px;",align="justify")
# merging = definemerging(df1,df2)
#### Cosmicinformation
cosmicinfo = merging[merging.COSMIC_ID.isna()==False].groupby(["gene_name","COSMIC_ID"]).count()
#### Cosmic ids
cinfo = cosmicinfo.reset_index()
cid = cinfo["COSMIC_ID"]
### Read info scrapped
cosbase = read_scrap()
cosmicdb = gendfclean(cosbase,cid)
cosmerge = merging.merge(cosmicdb,left_on="COSMIC_ID",right_on="cosmic_id",how="inner")
cols = ["tissue","histology"]
p1,p2 = st.columns((1,1))
for col,p in zip(cols,[p1,p2]):
dtemp = get_N_common(cosmerge,col)
#### Intermittently exposed
#### Continuously exposed
fig = px.bar(dtemp,x=dtemp.columns[0],y=dtemp.columns[1],color=dtemp.columns[2],barmode="group")
p.plotly_chart(fig)
### Percentage of VAF
wrt("2.11 Percentage of VAF captured in all mutations and its distribution. Depth Coverage distribution",tag="h4",align="left")
par=""" We want to proof that somatic mutations are tough to capture . To do that we have plot the variant allele frequency
distribution of the mutations to proof that the majority of them have a low score VAF
"""
wrt(par,tag="p style='font-size:21px;'",align="justify")
res = str(round(((df1.shape[0]-sum(df1['VAF']>0.05))/df1.shape[0])*100,2))+'%'
# plt.figure(figsize=(5,3))
fig = px.box(df1,y="VAF")
fig2 = px.box(df1,y="DP")
__,col1,__,col2,__ = st.columns((0.3,1,0.3,1,0.3))
# plt.text('Those outliers are the ones higher than 5%. There are',str(df1.shape[0]-sum(df1['VAF']<0.05)),'mutations higher than 5%.')
col1.plotly_chart(fig,clear_figure=True,use_container_width=True)
col2.plotly_chart(fig2,clear_figure=True,use_container_width=True)
par =f"""
Those outliers are the ones higher than 5%. There are {str(df1.shape[0]-sum(df1['VAF']<0.05))} mutations higher than 5%.
"""
wrt(par,tag="p",align="center")
wrt("2.12 Evolution of the C>T mutations with the age",tag="h4",align="left")
par =f"""
The plot shows us that the number of C>T (a UV associated mutation) mutations increases with the evolution of the age.
"""
wrt(par,tag="p style='font-size:21px;'",align="center")
plts = merging[merging.mut_type=="C>T"].groupby("age").count().reset_index()[["age","VAF"]]
plts.rename(columns={"VAF":"C>T"},inplace=True)
__,col1,__ = st.columns((1,3,1))
fig = plt.figure(5,2)
sns.lmplot(
data=plts,
x="C>T",
y="age",
logx=True,
height=3, aspect=3
).set(title="Logarithmic increase of number of C>T")
col1.pyplot()
wrt("2.13 Mean of pathogenic score in different genes",tag="h4",align="left")
# par =f"""
# The plot shows us that the number of C>T (a UV associated mutation) mutations increases with the evolution of the age.
# """
# wrt(par,tag="p style='font-size:21px;'",align="center")
# plts = merging[merging.mut_type=="C>T"].groupby("age").count().reset_index()[["age","VAF"]]
# plts.rename(columns={"VAF":"C>T"},inplace=True)
# __,col1,__ = st.columns((1,3,1))
# fig = plt.figure(5,2)
# sns.lmplot(data=plts,x="C>T",y="age",logx=True,height=3,aspect=3).set(title="Logarithmic increase of number of C>T")
# col1.pyplot()
dbinfo = merging.merge(cosmicdb,left_on="COSMIC_ID",right_on="cosmic_id",how="left")
df12 = dbinfo[dbinfo.cosmic_id.isna()==False].groupby("gene_name").mean().reset_index()[["gene_name","score"]].sort_values(by="score",ascending=False)
___,col,___ = st.columns((2,1,2))
with col:
st.dataframe(df12)
return None
def cidvars():
wrt("3. Individual gene information",tag="h2")
par1 = """
We are going to explore mutations of genes that occured within our samples as well as go beyond our data base, exploring
mutations that have cosmic ids so that we can show more detailed information about the mutation.
"""
wrt(par1,tag="p",align="center")
wrt("3.1 Where does mutations usually occur ?",tag="h3",align="left")
par = "In this section we visualize where the mutations usually occurs as well as tumours that may be caused these mutations. Note that we can get this information just for all the samples that contains cosmic id"
wrt(par,tag="p",align="justify")
merging = definemerging(df1,df2)
#### Cosmicinformation
cosmicinfo = merging[merging.COSMIC_ID.isna()==False].groupby(["gene_name","COSMIC_ID"]).count()
#### Cosmic ids
cinfo = cosmicinfo.reset_index()
cid = cinfo["COSMIC_ID"]
### Read info scrapped
cosbase = read_scrap()
cosmicdb = gendfclean(cosbase,cid)
cosmerge = merging.merge(cosmicdb,left_on="COSMIC_ID",right_on="cosmic_id",how="inner")
cols = ["tissue","histology"]
p1,p2 = st.columns((1,1))
for col,p in zip(cols,[p1,p2]):
dtemp = get_N_common(cosmerge,col)
#### Intermittently exposed
#### Continuously exposed
fig = px.bar(dtemp,x=dtemp.columns[0],y=dtemp.columns[1],color=dtemp.columns[2],barmode="group")
p.plotly_chart(fig)
### Mut type cus plot
wrt("3.2 How is the mutation in comparison with the reference base ?",tag="h3",align="left")
par ="""
We want to know between all the mutations occured (with and without cosmic id) what are the most common mutation type occured
"""
wrt(par,tag="p",align="justify")
merging["mut_type_cus"] = merging.apply(mut_type,axis=1)
plot4 = merging.groupby(["UV_exposure_tissue","mut_type_cus"]).count().reset_index().iloc[:,:3]
plot4 = plot4.rename(columns={"sampleID":"n_mut"})
fig = px.bar(plot4,x="mut_type_cus",y="n_mut",color="UV_exposure_tissue",barmode="group")
col1.plotly_chart(fig,use_container_width=True)
col2.plotly_chart()
return None
## Slide in which we will explain the pipeline followed in order to analyze the two samples we have.
def set_igv():
wrt("3. Sample Analysis (IGV)",tag="h2")
par1 = """
Explanation of the steps followed to process our raw samples into files which can be interpreted with the IGV program (VCF files).
"""
wrt(par1,tag="p style='font-size:21px;'",align="center")
wrt("3.1 Which Samples do we have?",tag="h4",align="left")
par2 = """
We wanted two samples of very oposite individuals, with the goal of being able to appreciate noticable differences when sequencing
the genes.
- AG0312: Sample of a Male individual of 70 years old, whose skin sample is Intermittently-photoexposed, and it has a skin phototype of I.
- AG0322: Sample of a Female individual of 38 years old, whose skin sample is Chronically-photoexposed, and it has a skin phototype of IV.
"""
wrt(par2,tag="p style='font-size:21px;'",align="justify")
wrt("3.2 Processing Steps",tag="h4",align="left")
par3 = """
- 1. Quality Control and Alignment
Initially, both samples were in bam format, thus are samples of good quality (since they passed the quality step) and also are aligned.
Thus we started our preprocessing by refinning the Alignment.
"""
par4 = """
- 2. Refinement of Alignment
Since for performing Variant Calling we need to have our samples sorted by genomic positions. Once sorted we mark
the duplicates (in these exist) so that the these are ignored in the Variant Calling step.
Finally we just index the resulting bam files for its future visualization.
"""
par5 = """
- 3. Variant Calling
Once we arrived to this section we had to solve an error which initially we didn't knew why it happend. The error ocurred when
we tried to identify the active regions (error shown on the below image).
The problem consited on that, in the refined bam the chromosomes were represented just with their number, and in the reference
genome (hg19) chromosomes were represented as "chr N". Thus just by eliminating the "chr" part in the reference genome, the
variant calling could be accomplished.
"""
par6 = """
- 4. Visualization with IGV
With the variant calling files computed, we are ready now to display the variants in our Alignments.
We tried to do a simple task, which was just to try to find the mutations which the researchers of the paper found for our samples.
For doing this we used the chromosome and position indicators of their dataset.
Near the positions idicated we found the mentioned mutation in the dataset, however not in the exact position.
"""
wrt(par3,tag="p style='font-size:21px;'",align="justify")
wrt(par4,tag="p style='font-size:21px;'",align="justify")
wrt(par5,tag="p style='font-size:21px;'",align="justify")
st.image("img/error.png",use_column_width=True)
wrt(par6,tag="p style='font-size:21px;'",align="justify")
st.image("img/chartc.png",use_column_width=True)
st.header("\n")
st.header("\n")
st.header("\n")
wrt("Igv Images",tag="h1",align="center")
st.header("\n")
st.header("\n")
st.header("\n")
__,col1,__,col2,__ = st.columns((0.2,1.5,0.3,1,0.2))
with col1:
st.image("img/col1igv.png",use_column_width=True,caption="Fig 5: Igv rep AG0312")
with col2:
st.image("img/col2igv.png",use_column_width=True,caption="Fig 6: Igv representation AG0322")
return None
if menu == '1. Intro':
set_home()
elif menu == '2. Analysis of somatic mutations':
set_chintp()
# elif menu == '3. Individual sample information':
# cidvars()
elif menu == '3. Sample Analysis (IGV)':
set_igv()
# elif menu == 'Otras variables':
# set_otras_variables()
# elif menu == 'Relaciones entre variables':
# set_relations()
# elif menu == 'Matrices de correlación':
# set_arrays() |