Example to convert excel corrupted human gene names back from dates to gene symbols

date2gene.py

#!/usr/bin/env python3
import os
import warnings
import re
import pandas as pd

date_re = re.compile(r'\d{4}-\d{2}-\d{2}')
mnth2str = {3: 'MARCH', 9: 'SEPT', 10: 'OCT', 12: 'DEC'}


def date2gene(x):
    try:
        mnth = mnth2str[pd.to_datetime(x).month]
        day = pd.to_datetime(x).day
        return "{}{}".format(mnth, day)
    except KeyError:
        warnings.warn("Could not convert date: {}".format(x))
        return x


def convert_column(df, column):
    '''
    Convert dates in column to gene names.

    Args:
        df (pandas.DataFrame):
            dataframe to convert

        column (str):
            name of column with gene names to convert
    '''
    df[column] = df[column].apply(lambda x: date2gene(x)
                                  if date_re.match(str(x)) else x)


xlsx = '/path/to/file.xlsx'  # edit as appropriate
sheet_name = 'my_gene_list'  # edit as appropriate
column = 'gene_sybols'  # edit as appropriate
df = pd.read_excel(xlsx, sheet_name=sheet_name)
convert_column(df, column)
f, ext = os.path.splitext(xlsx)
out = f + '.date2gene.xlsx'
df.to_excel(out, sheet_name=sheet_name, index=False)