# -*- coding: utf-8 -*-
"""Main module."""
import sys
import click
import pandas as pd
from pimkl.network import get_network_from_csv
from pimkl.inducers import (
read_gmt, get_matching_data_and_network, write_preprocessed
)
[docs]def invalid_name(name):
return '_' in name or '-' in name
[docs]def assert_valid_names(*names):
for name in names:
if invalid_name(name):
raise IOError(
'names cannot contain "_" or "-", correct "{}"'.format(name)
)
[docs]def reader(filename):
return pd.read_csv(filename, sep=',', index_col=0).T
[docs]def read_data(filename, reader, gene_name_transformation=None):
data = reader(filename)
data.index.name = 'samples'
data.columns.name = 'gene_name'
if gene_name_transformation is not None:
data.columns = [
gene_name_transformation(gene_name) for gene_name in data.columns
]
return data
[docs]def preprocess_data_and_inducers(
data_csv_files,
data_names,
network_csv_file,
network_name,
gene_sets_gmt_file,
gene_sets_name,
preprocess_dir,
match_samples
):
"""
Inducers, that is Laplacian matrices for geneset subnetworks, and data are
preprocessed and written to file.
Data and inducers are filtered for genes (per dataset) available in the
data and the network and the union of genesets.
Conditionally, also the data is filtered for matching samples over all
datasets.
"""
assert_valid_names(*data_names, network_name, gene_sets_name)
data_dict = {
data_name: read_data(filepath, reader)
for data_name, filepath in zip(data_names, data_csv_files)
# TODO test data no list
}
# filter for matching samples
if match_samples:
samples_set = None
for data_name, data in data_dict.items():
print(data.shape)
if samples_set is None:
samples_set = set(data.index)
else:
samples_set &= set(data.index)
print(data_name, data.shape)
matching_samples = sorted(list(samples_set))
for _, data in data_dict.items():
data = data.loc[matching_samples]
print(data.shape)
# filter for matching genes
gene_sets = read_gmt(gmt_file=gene_sets_gmt_file)
all_genes_from_gene_sets = set()
for set_name, genes in gene_sets.items():
all_genes_from_gene_sets |= genes
considered_genes = list(all_genes_from_gene_sets)
network = get_network_from_csv(network_csv_file,
sep=',').get_sub_network(considered_genes)
# write data and inducers
for data_name, data in data_dict.items():
data_filtered, network_filtered = get_matching_data_and_network(
data, network
)
write_preprocessed(
data=data_filtered,
data_name=data_name,
network=network_filtered,
network_name=network_name,
gene_sets=gene_sets,
gene_sets_name=gene_sets_name,
output_dir=preprocess_dir
)
@click.command()
@click.option('-fd', '--data_csv_file', 'data_csv_files', required=True, multiple=True, type=click.Path())
@click.option('-nd', '--data_name', 'data_names', required=True, multiple=True)
@click.argument('network_csv_file', required=True, type=click.Path())
@click.argument('network_name', required=True)
@click.argument('gene_sets_gmt_file', required=True, type=click.Path())
@click.argument('gene_sets_name', required=True)
@click.argument('preprocess_dir', required=True, type=click.Path(exists=True, file_okay=False, writable=True))
def main(
data_csv_files,
data_names,
network_csv_file,
network_name,
gene_sets_gmt_file,
gene_sets_name,
preprocess_dir,
match_samples=True
):
"""
Compute incuding Laplacian matrices and preprocess data matrices for
matching features.
Multiple data_csv_files may be passed.
Each data_csv_file should readable as pandas.DataFrames
`pd.read_csv(filename, sep=',', index_col=0)` where index are features
(rows) and columns a are samples.
The `network_csv_file` is an edge list readable with `pd.read_csv(filename)`
where the 3rd columns is a numeric value.
The `gene_sets_gmt_file` should follow the gmt specification. See
http://software.broadinstitute.org/cancer/software/gsea/wiki/index.php/Data_formats
For each file, a name has to be passed. Names cannot contain "_" or "-".
Results are written to `preprocess_dir`.
"""
preprocess_data_and_inducers(
data_csv_files,
data_names,
network_csv_file,
network_name,
gene_sets_gmt_file,
gene_sets_name,
preprocess_dir,
match_samples
)
return 0
if __name__ == "__main__":
sys.exit(main()) # pragma: no cover