pandas dataframe的一个技巧

import sys,os,re
import numpy as np
import pandas as pd

from glob import glob


#实际路径：/mnt/lustre/user/wubin/03.Clinical_Data/16.EQA/2022_solid_tumor/no_split
os.chdir('M:/')

file_lst = sorted(glob('*.all.snv_indel.eqa.HGVS.xls'))

df_all = pd.DataFrame() #定义一个空的DataFrame

for file in file_lst:
    df = pd.read_csv(file,sep='\t')
    sample = file.split('.')[0]
    sample = re.sub('^22SSP0', '', sample)
    df.insert(0, '样本编号', sample)
    df_all = pd.concat([df_all,df],ignore_index=True)
    

df_all.to_excel('E:/XXXX/SNV_INDEL.xlsx',
                index=False,
                freeze_panes=(1,1))

df_all = pd.DataFrame() #定义一个空的DataFrame

这个用法很好，连columns都不用预先定义