updated schema

allowed for `published_auc` as another dose response value

Author: sgosline

build/pancpdo/04-getPancPDOExperiments.py

@@ -44,6 +44,7 @@ def map_to_drugs_samps(dose_rep,drugfile,sampfile):
     print(merged)
     return merged
 
+
 def get_data(token):
     synid = 'syn64333325'
 

build/pancpdo/05-addPrecalcAUC.py

@@ -0,0 +1,50 @@
+
+
+
+import os
+import pandas as pd
+import wget
+import argparse
+import synapseclient as sc
+import math
+import re
+
+
+def get_precalc_auc():
+    '''
+    get pre-calculaterd AUC from supp data
+    '''
+    tablink = 'https://aacr.silverchair-cdn.com/aacr/content_public/journal/cancerdiscovery/8/9/10.1158_2159-8290.cd-18-0349/5/21598290cd180349-sup-199398_2_supp_4775187_p95dln.xlsx?Expires=1738004990&Signature=av8XadTm9AmI20O2Y7J7aHDtPbpluKJIfI5ubsoiYJ15D0zh5p1ltF4a7-DCSWTSMs-qX5TD09shxHeqkQ2NkLWHZsXoCD5KyREGhEgcDAvWZ1V9kwXDm0bjpINipAPPtC20oeuw6c~hPooF3Mtgzp4MzMCCjcVwfn05u27a0kS0yifBi11wQj3nmHlR3ym-2fYkFuqQtnNPCzH8-yIw21y0kTvXrNodAzC5pGA8qUK4PLxBt52xUIvTEPsPiPjXwBnDCfVsLGGdDYIY25lEPKiA403q6kFYvrSQ3bsTvM4kuvltb7yS4AXjK0-tthMOKbqq8~uREmJCcueADUF91g__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA'
+
+    chemo = pd.read_excel(tablink,sheet_name=1,skiprows=1)
+    chemo.columns = [c.lower() for c in chemo.columns]
+    targeted = res = pd.read_excel(tablink,sheet_name=2,skiprows=1)
+    targeted.columns = [c.lower() for c in targeted.columns]
+
+    cdat = chemo.melt(id_vars='sample id',value_vars=['gemcitabine','paclitaxel','sn-38','5-fu','oxaliplatin'],var_name='drug',value_name='published_auc')
+    tdat = targeted.melt(id_vars='sample id',value_vars=set(targeted.columns)-set('sample id'),var_name='drug',value_name='published_auc')
+    combined= pd.concat([cdat,tdat])
+    combined = combined.rename(columns={'sample id':'other_id','drug':'chem_name'})
+    
+    return combined
+
+def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument('-s', '--samples', help='Sample mapping file for panc pdo samples')
+    parser.add_argument('-d', '--drugs', help='Drug mapping file for panc pdo samples')
+    parser.add_argument('-e', '--expfile', default = '/tmp/pancpdo_experiments.tsv',help='Output file to be read into curve fitting code')
+
+    args = parser.parse_args()
+    samples = pd.read_csv(args.samples,sep=',')
+    drugs = pd.read_tsv(args.drugs,sep='\t')
+
+    newdat = get_precalc_auc().merge(samples).merge(drugs)
+    newdat = newdat[['improve_sample_id','improve_drug_id','published_auc']].drop_duplicates()
+    newdat = newdat.melt(id_vars=['improve_sample_id','improve_drug_id'],value_vars='published_auc',var_name='dose_response_metric',value_name='dose_response_value')
+    newdat[['source']]='TiriacEtAl2018'
+    newdat[['time']]=120
+    newdat[['time_unit']]='hours'
+    newdat[['study']]='pancpdo'
+    oldat = pd.read_csv(args.expfile,sep='\t')
+    res = pd.concat([olddat,newdat])
+    res.to_csv(args.expfile)

build/pancpdo/build_exp.sh

@@ -5,8 +5,10 @@ trap 'echo "Error on or near line $LINENO while executing: $BASH_COMMAND"; exit
 
 echo "Running 04-drug_dosage_and_curves.py with drugfile $2 and curSampleFile $1"
 python 04-getPancPDOExperiments.py --pat $SYNAPSE_AUTH_TOKEN --drugs $2 --samples $1 --output /tmp/pancpdo_doserep.tsv
-python fit_curve.py --input /tmp/pancpdo_doserep.tsv
+python fit_curve.py --input /tmp/pancpdo_doserep.tsv --output /tmp/pancpdo_doserep.tsv
 
 ##now move file and gzip
-mv /tmp/pancpdo_doserep.tsv /tmp/pancpdo_experiments.tsv
+mv /tmp/pancpdo_doserep.tsv.0 /tmp/pancpdo_experiments.tsv
+
+python 05-addPrecalcAUC.py --samples $1 --drugs $2 --expfile /tmp/pancpdo_experiments.tsv
 gzip /tmp/pancpdo_experiments.tsv

schema/coderdata.yaml

@@ -221,6 +221,8 @@ enums:
         description: I believe this is the drug sensitivity score
       mrecist:
         description: For PDX data this value should be either Progressive Disease, Stable Disease, Partial Response or Complete Response.
+      published_auc:
+        description: AUC that has been published with the study. Currently used in organoid data.
       gc_auc:
         description: Area under tumor growth curve
       efs: