python处理tsv文件

Question

现有一个矩阵（tsv文件），行名为ENSG开头+数字编号
前五行五列情况如图

想要读取output.json文件中如图的对应关系，把ENSG开头+数字编号替换为后面的代号内容
例如：行名为ENSG00000186092 替换为OR4F5

其中output.json中各个列使用\t分隔
输出名称替换后的矩阵（tsv文件）到指定路径

Answer 1

    保留在json中不存在的数据：
    import pandas as pd
    json_data = pd.read_json("xxx.json", typ='series')
    json_dict = json_data.to_dict()
    df_chunk = pd.read_csv("xxx.tsv", sep='\t', chunksize=1000)
    df_chunk_list = []
    for chunk in df_chunk:
        chunk['Ensembl_ID'] = chunk['Ensembl_ID'].apply(lambda x: x.split(".")[0])
        for index, row in chunk.iterrows():
            try:
                chunk.loc[index, 'Ensembl_ID'] = json_dict[row['Ensembl_ID']]
            except:
                pass
        df_chunk_list.append(chunk)
    result_Df = pd.concat(df_chunk_list)
    result_Df.to_csv('result.tsv', sep='\t', index=False)

    不保留在json中不存在的数据：
    import pandas as pd
    json_data = pd.read_json("xxx..json", typ='series')
    json_dict = json_data.to_dict()
    df_chunk = pd.read_csv("xxx..tsv", sep='\t', chunksize=1000)
    df_chunk_list = []
    for i, chunk in enumerate(df_chunk):
        chunk['Ensembl_ID'] = chunk['Ensembl_ID'].apply(lambda x: x.split(".")[0])
        for index, row in chunk.iterrows():
            try:
                chunk.loc[index, 'Ensembl_ID'] = json_dict[row['Ensembl_ID']]
            except:
                chunk.drop(index=[index], inplace=True)
        df_chunk_list.append(chunk)
    result_Df = pd.concat(df_chunk_list)
    result_Df.to_csv('result.tsv', sep='\t', index=False)