[Py009] pandas根据指定列合并其他列内容

从网上下载到lncRNA的疾病信息

现在需要根据"Gene Symbol"列整合其对应的“Alias","Disease","PMID"，如下图

image

处理过程如下：

import pandas as pd
disease = pd.read_table("lncbook_disease.txt", sep="\t")
# piece = dict(list(disease.groupby('Gene Symbol')))

def rows_content_joined(df, spr, *cols):
    tmp = pd.DataFrame()
    for i in cols:
        try:
            content = spr.join(set(df[i].tolist()))
        except TypeError:
            content = spr.join(set(df[i].apply(str).tolist()))  # 处理数值型的列
        tmp[i] = pd.Series(content)
    return tmp

results = pd.DataFrame()
for name, group in disease.groupby('Gene Symbol'):
    results = results.append(rows_content_joined(group, "; ", "Gene Symbol", "Alias", "Disease", "PMID"))

results.to_csv("lncbook_disease_handled.txt", header=True, index=False, sep="\t")