Mercurial > repos > mvdbeek > damidseq_consecutive_peaks

comparison consecutive_peaks.py @ 0:7f827a8e4ec5 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/bardin-lab/damid_galaxy_tools commit c753dd4f3e1863aae7ba45dcc7efdf6937b03542-dirty
author mvdbeek
date Fri, 26 Oct 2018 11:58:06 -0400
parents
children f3ca59e53b73
comparison
equal deleted inserted replaced
-1:000000000000 0:7f827a8e4ec5
1 import click
2 import numpy as np
3 import pandas as pd
4
5 SHIFTED_PADJ_COLUMN = 'shifted'
6 CONSECUTIVE_MAX = 'consecutive_max'
7 PEAKS_PER_GROUP = 'peaks_per_group'
8
9
10 @click.command()
11 @click.argument('input_file', type=click.Path(exists=True))
12 @click.argument('output_file', type=click.Path())
13 @click.argument('padj_column', default=8)
14 @click.argument('groupby_column', default=9)
15 @click.argument('add_number_of_peaks', default=True)
16 def determine_consecutive_peaks(input_file, output_file, padj_column, groupby_column, add_number_of_peaks):
17 """Finds the two lowest consecutives peaks for a group and reports"""
18 df = pd.read_csv(input_file, sep='\t', header=None)
19 grouped = df.groupby(groupby_column, sort=False)
20 if add_number_of_peaks:
21 df[PEAKS_PER_GROUP] = grouped[groupby_column].transform(np.size)
22 df[SHIFTED_PADJ_COLUMN] = grouped[8].shift()
23 df[CONSECUTIVE_MAX] = df[[padj_column, SHIFTED_PADJ_COLUMN]].max(axis=1)
24 grouped = df.groupby(groupby_column, sort=False)
25 idx = grouped[CONSECUTIVE_MAX].transform(min) # index of groupwise consecutive minimum
26 new_df = df[df[CONSECUTIVE_MAX] == idx]
27 new_df.sort_values(by=CONSECUTIVE_MAX)
28 new_df[padj_column].replace(new_df[CONSECUTIVE_MAX])
29 new_df = new_df.drop(labels=[CONSECUTIVE_MAX, SHIFTED_PADJ_COLUMN], axis=1)
30 new_df.to_csv(output_file, sep='\t', header=None, na_rep="NaN")
31
32
33 if __name__ == '__main__':
34 determine_consecutive_peaks()