Correct a TCGA assembled tsv file (tab delimited), formatting sample names for tsv columns

tcga_correct_samplenames.py

import pandas as pd
import sys
import re

tcga_tsv = sys.argv[1]

tcga = pd.read_table(tcga_tsv, sep='\t', index_col=0)

oldcolumns = tcga.columns.tolist()
newcolumns = ['-'.join(re.findall(r'TCGA[^_]*', oc)[0].split('-')[:4])
              for oc in oldcolumns]

tcga.columns = newcolumns
tcga.to_csv(tcga_tsv, sep='\t')