def find_adjacencies_command(args):
"""
Infer co-expression modules.
"""
LOGGER.info("Loading expression matrix.")
ex_mtx = _load_expression_matrix(args)
tf_names = load_tf_names(args.tfs_fname.name)
n_total_genes = len(ex_mtx.columns)
n_matching_genes = len(ex_mtx.columns.isin(tf_names))
if n_total_genes == 0:
LOGGER.error(
"The expression matrix supplied does not contain any genes. Make sure the extension of the file matches the format (tab separation for TSV and comma sepatration for CSV)."
)
sys.exit(1)
if float(n_matching_genes) / n_total_genes < 0.80:
LOGGER.warning(
"Expression data is available for less than 80% of the supplied transcription factors."
)
LOGGER.info("Inferring regulatory networks.")
client, shutdown_callback = _prepare_client(args.client_or_address,
num_workers=args.num_workers)
try:
network = grnboost2(expression_data=ex_mtx,
tf_names=tf_names,
verbose=True,
client_or_address=client)
finally:
shutdown_callback(False)
LOGGER.info("Writing results to file.")
network.to_csv(args.output, index=False, sep='\t')
def find_adjacencies_command(args):
"""
Infer co-expression modules.
"""
LOGGER.info("Loading expression matrix.")
try:
ex_mtx = load_exp_matrix(
args.expression_mtx_fname.name,
(args.transpose == 'yes'),
args.sparse,
args.cell_id_attribute,
args.gene_attribute,
)
except ValueError as e:
LOGGER.error(e)
sys.exit(1)
tf_names = load_tf_names(args.tfs_fname.name)
if args.sparse:
n_total_genes = len(ex_mtx[1])
n_matching_genes = len(ex_mtx[1].isin(tf_names))
else:
n_total_genes = len(ex_mtx.columns)
n_matching_genes = len(ex_mtx.columns.isin(tf_names))
if n_total_genes == 0:
LOGGER.error(
"The expression matrix supplied does not contain any genes. "
"Make sure the extension of the file matches the format (tab separation for TSV and "
"comma sepatration for CSV)."
)
sys.exit(1)
if float(n_matching_genes) / n_total_genes < 0.80:
LOGGER.warning("Expression data is available for less than 80% of the supplied transcription factors.")
LOGGER.info("Inferring regulatory networks.")
client, shutdown_callback = _prepare_client(args.client_or_address, num_workers=args.num_workers)
method = grnboost2 if args.method == 'grnboost2' else genie3
try:
if args.sparse:
network = method(
expression_data=ex_mtx[0],
gene_names=ex_mtx[1],
tf_names=tf_names,
verbose=True,
client_or_address=client,
seed=args.seed,
)
else:
network = method(
expression_data=ex_mtx, tf_names=tf_names, verbose=True, client_or_address=client, seed=args.seed
)
finally:
shutdown_callback(False)
LOGGER.info("Writing results to file.")
extension = PurePath(args.output.name).suffixes
network.to_csv(args.output.name, index=False, sep=suffixes_to_separator(extension))
