study.py

import pandas as pd
from functools import partial
from collections import Iterable

STORE_DIR = "preprocessed"

def do_df(filename, sep=" ", new_name=None, do_transpose=False, plugin_func=None):
    df = pd.read_table(filename, sep=sep, index_col=0)
    if do_transpose:
        df = df.T

    if plugin_func:
        try:
            for func in plugin_func:
                df = func(df)
        except:
            df = plugin_func(df)

    if new_name:
        df.to_csv("%s/%s" % (STORE_DIR, new_name), sep=" ")
    return df

def skip_lines(df, rows=None, cols=None):
    wdf = df.copy()
    if rows:
        wdf = wdf.drop(wdf.index[rows])
    if cols:
        wdf = wdf.drop(wdf.columns[cols], 1)

    return wdf


def columns_to_upper(df):
    df.columns = [col.upper()  for col in df.columns]
    return df


def main():
    do_df("SampleCharGCT.csv", sep=",", new_name="phenotype.csv")
    do_df("GoldenGateMethGCT.txt", sep="\t", new_name="meth.csv", do_transpose=True,
          plugin_func=partial(skip_lines, rows=[0, 1]))
    do_df("score_matrix_tarbase_GCT_ext.txt", sep="\t", new_name="interaction_TAR.csv",
          plugin_func=columns_to_upper)
    do_df("score_matrix_mirwalk_GCT_05.txt", sep="\t", new_name="interaction_MIR.csv",
          plugin_func=columns_to_upper)

    do_df("StemCellqPCRGCTNormalized.txt", sep=" ", new_name="mRNA.csv", do_transpose=True)
    do_df("miRNANormalizedGCTnames.txt", sep=" ", new_name="miRNA.csv", do_transpose=True)


if __name__ == "__main__":
    main()