简单的数据处理代码

代码分享
#!/usr/bin/env python3

'''

used for CountSummarizer, DataFetcher and TaxonomyAdder
auhtor, Xun
last modified, 2023-07-07
Mail, makise.kurisu.yoru@gmail.com

'''

import pandas as pd
import argparse


class CountSummarizer:
    def __init__(self, args):
        self.args = args

    def norm_expr(self, probe_expr_df, norm_type):
        if norm_type == "raw":
            return probe_expr_df
        elif norm_type == "cpm":
            norm_factor = probe_expr_df.sum()
            return probe_expr_df * 1000000 / norm_factor

    def run(self):
        args = self.args
        probe_expr_df = pd.read_csv(args.input_file, sep="\t", header=0, index_col=0)
        probe_expr_df.index = probe_expr_df.index.map(str)

        # Normalize expression data
        abundances = [self.norm_expr(probe_expr_df, norm_type) for norm_type in args.norm_type]

        usecols = [0] + args.grp_column
        probe_map_df = pd.read_csv(args.map_file, sep="\t", header=0, index_col=None, dtype=str, usecols=usecols)
        key_column = probe_map_df.columns[0]
        
        # Modified part starts here
        grp_columns = probe_map_df.columns[1:]
        
        # Ensure the separators match the columns
        if len(args.grp_sep) == 1:
            grp_seps = args.grp_sep * len(grp_columns)
        else:
            grp_seps = args.grp_sep
        
        for grp_column, grp_sep in zip(grp_columns, grp_seps):
            self.process_grp_column(grp_column, grp_sep, key_column, abundances, args.norm_type, args.abundance_keep, probe_map_df)
        # Modified part ends here

    def process_grp_column(self, grp_column, grp_sep, key_column, abundances, norm_type, abundance_keep, probe_map_df):
        """Processes each group column and saves the aggregated data."""
        # Extract the DataFrame for the group column

        grp_df = probe_map_df.loc[probe_map_df[grp_column] != "", [key_column, grp_column]]
        args = self.args
        if grp_sep == "*":
            grp_sep = ""
        grp_df[grp_column] = grp_df[grp_column].str.split(grp_sep)
        grp_df_explode = grp_df.explode(grp_column)
        grp_df_explode = grp_df_explode.loc[grp_df_explode[grp_column] != "", ]

        # Iterate through normalized abundance dataframes
        for abundance_df, norm_type in zip(abundances, norm_type):
            merged_df = grp_df_explode.merge(abundance_df, left_on=key_column, right_index=True, how="inner")

            # Aggregating data
            aggregation_dict = {col: abundance_keep for col in merged_df.columns if col not in [grp_column, key_column]}
            final_df = merged_df.groupby(grp_column).agg(aggregation_dict).reset_index()

            final_df.to_csv(f"{args.output_prefix}.{grp_column}.{norm_type}.txt", sep="\t", index=False)




class DataFetcher:
    def __init__(self, args):
        self.args = args

    def fetch_data(self):
        args = self.args
        df_input = pd.read_table(args.input, sep='\t', header=None, dtype=str)
        df_reference = pd.read_table(args.reference, sep='\t', header=None, dtype=str)

        # Convert columns to integers for matching
        col1 = int(args.column.split(',')[0]) - 1
        col2 = int(args.column.split(',')[1]) - 1
        list_input = df_input.iloc[:, col1].tolist()
        list_reference = df_reference.iloc[:, col2].tolist()

        # Find common elements
        common_elements = set(list_input) & set(list_reference)

        # Filter the input DataFrame to only include rows with common elements
        df_filtered_input = df_input[df_input.iloc[:, col1].isin(common_elements)]
        
        # Filter the reference DataFrame to only include rows with common elements
        df_filtered_ref = df_reference[df_reference.iloc[:, col2].isin(common_elements)]

        # Save the output
        if args.show_other_columns:
            df_output = pd.merge(df_filtered_input, df_filtered_ref, left_on=col1, right_on=col2, how='inner')
        else:
            df_output = df_filtered_input

        df_output.to_csv(args.output, index=False, sep='\t', header=None)


def create_parser():
    parser = argparse.ArgumentParser(prog='anno_db.py', description='anno_db includes two main functionalities: data fetching with "cr" command and abundance summarizing with "ex" command.')
    subparsers = parser.add_subparsers(dest='command')

    # Subparser for the 'cr' command
    fetch_parser = subparsers.add_parser('cr', description='This command fetches data from the input file and reference file based on a given column, and outputs the matched rows to the output file.')
    fetch_parser.add_argument("-i", "--input", help="The input file.")
    fetch_parser.add_argument("-r", "--reference", help="The reference file.")
    fetch_parser.add_argument("-o", "--output", help="The output file where the result will be saved.")
    fetch_parser.add_argument("-c", "--column", default="1,1", help="The column indices in the input and reference files for matching rows, separated by a comma. Indexing starts from 1.")
    fetch_parser.add_argument("-s", "--show_other_columns", action="store_true", help="Whether to show other columns in the reference file.")

    # Subparser for the 'ex' command
    ex_parser = subparsers.add_parser('ex', description='This command executes the abundance summarization on the input file based on the group information in the map file.')
    ex_parser.add_argument("-i", "--input_file", help="Sub-item abundance file with format specified above", required=True)
    ex_parser.add_argument("-m", "--map_file", help="Map file containing group information", required=True)
    ex_parser.add_argument("-e", "--abundance_keep", default="sum", choices=['sum', 'median', 'min', 'max', 'mean'], help="Keep abundance as median, sum, min, max, or mean")
    ex_parser.add_argument("-c", "--grp_column", default=[2], type=lambda s: [int(item) - 1 for item in s.split(',') if item], help="The column(s) contains group information")
    ex_parser.add_argument("-s", "--grp_sep", default=[","], type=lambda s: s.split('+'), help="Separator(s) for each group")
    ex_parser.add_argument("-n", "--norm_type", default=["raw","cpm"], type=lambda s: s.split(','), help="Specify the output data type")
    ex_parser.add_argument("-o", "--output_prefix", default="output", help="Output file prefix")

    # Subparser for the 'tax' command
    tax_parser = subparsers.add_parser('tax', description='This command adds taxonomy information to the cds.output file.')
    tax_parser.add_argument("-i", "--cds_output_file", help="The cds.output file.")
    tax_parser.add_argument("-t", "--taxonomy_file", help="The taxonomy.tsv file.")
    tax_parser.add_argument("-o", "--output_dir", help="The output directory.")

    return parser

def main():
    parser = create_parser()
    args = parser.parse_args()

    if args.command == 'cr':
        fetcher = DataFetcher(args)
        fetcher.fetch_data()
    elif args.command == 'ex':
        summarizer = CountSummarizer(args)
        summarizer.run()
    else:
        print("Invalid command")

if __name__ == "__main__":
    main()





该代码用于数据处理和丰度汇总,详细见帮助
我也发了一个大概用法的介绍,可以看一下
https://www.omicsclass.com/article/2337

  • 发表于 2023-10-25 11:54
  • 阅读 ( 731 )
  • 分类:python

你可能感兴趣的文章

相关问题

0 条评论

请先 登录 后评论
xun
xun

电路元件工程师

82 篇文章

作家榜 »

  1. omicsgene 698 文章
  2. 安生水 347 文章
  3. Daitoue 167 文章
  4. 生物女学霸 120 文章
  5. xun 82 文章
  6. 红橙子 78 文章
  7. rzx 74 文章
  8. CORNERSTONE 72 文章