Coverts postProcessed files to have entrez IDs.

ConvertGS2Entrez_postProc.py

import csv 
import pandas as pd

gene2entrez = []
with open('gene2entrez.csv', mode='r') as infile:
    reader1 = csv.reader(infile)
    gene2entrez = {row1[0]:row1[1] for row1 in reader1}

for  post in ['pita','targetscan','tfbs_db']:       
    # read in the gene symbol map and the post proc
    pp1 = pd.read_csv( 'postProcessed_vSurv/postProcessed_GBM_'+post+'.csv', header = 0, index_col = 0)
    
    # tfsplit the post proc
    ppcols_expMatches = ['TFBS_DB.Exapnded Matches','Up.WEEDER Motif2 Expanded Matches','Up.WEEDER Motif1 Expanded Matches','Up.MEME Motif2 Expanded Matches','Up.MEME Motif1 Expanded Matches']
    for ppcol in ppcols_expMatches:
        for pprow in pp1.index.values:
            if not pd.isnull(pp1.loc[pprow,ppcol]):
                print pp1.loc[pprow,ppcol]
                tmp1 = ' '.join([':'.join([gene2entrez[i.split(':')[0]],';'.join([gene2entrez[j] for j in i.split(':')[1].split(';')])]) for i in pp1.loc[pprow,ppcol].split(' ')])
                print tmp1
                pp1.loc[pprow,ppcol] = tmp1

    ppcols_corMatches = ['TFBS_DB.Minimum Correlated_GBM','TFBS_DB.Correlated Matches_GBM','Up.WEEDER Motif2 Minimum Correlated_GBM','Up.WEEDER Motif2 Correlated Matches_GBM','Up.WEEDER Motif1 Minimum Correlated_GBM','Up.WEEDER Motif1 Correlated Matches_GBM','Up.MEME Motif2 Minimum Correlated','Up.MEME Motif2 Correlated Matches_GBM','Up.MEME Motif1 Minimum Correlated_GBM','Up.MEME Motif1 Correlated Matches_GBM']
    for ppcol in ppcols_corMatches:
        for pprow in pp1.index.values:
            if not pd.isnull(pp1.loc[pprow,ppcol]):
                tmp1 = ' '.join([gene2entrez[i.split(':')[0]]+':'+i.split(':')[1]+':'+i.split(':')[2] for i in pp1.loc[pprow,ppcol].split(' ')])
                print tmp1
                pp1.loc[pprow,ppcol] = tmp1

    pp1.fillna('NA').to_csv('postProcessed_vSurv/postProcessed_GBM_Entrez_'+post+'.csv')