def gmt2regions(gmt_fname, db_fname, delineation_code, fraction):
db = RegionRankingDatabase(fname=db_fname, name=os.path.basename(db_fname))
signatures = GeneSignature.from_gmt(gmt_fname,
nomenclature=db.nomenclature)
delineation = CODE2DELINEATION[delineation_code]
for signature in signatures:
sys.stdout(
signature.name + ',' +
','.join(convert(signature, db, delineation, fraction).genes))
def load_modules(fname: str) -> Sequence[Type[GeneSignature]]:
# Loading from YAML is extremely slow. Therefore this is a potential performance improvement.
# Potential improvements are switching to JSON or to use a CLoader:
# https://stackoverflow.com/questions/27743711/can-i-speedup-yaml
# The alternative for which was opted in the end is binary pickling.
extension = PurePath(fname).suffixes
if is_valid_suffix(extension, 'ctx_yaml'):
return load_from_yaml(fname)
elif '.dat' in extension:
with openfile(fname, 'rb') as f:
return pickle.load(f)
elif '.gmt' in extension:
return GeneSignature.from_gmt(fname)
else:
raise ValueError("Unknown file format for \"{}\".".format(fname))
def load_modules(fname: str) -> Sequence[Type[GeneSignature]]:
# Loading from YAML is extremely slow. Therefore this is a potential performance improvement.
# Potential improvements are switching to JSON or to use a CLoader:
# https://stackoverflow.com/questions/27743711/can-i-speedup-yaml
# The alternative for which was opted in the end is binary pickling.
if fname.endswith('.yaml') or fname.endswith('.yml'):
return load_from_yaml(fname)
elif fname.endswith('.dat'):
with open(fname, 'rb') as f:
return pickle.load(f)
elif fname.endswith('.gmt'):
sep = guess_separator(fname)
return GeneSignature.from_gmt(fname,
field_separator=sep,
gene_separator=sep)
else:
raise ValueError("Unknown file format for \"{}\".".format(fname))
def aucell_command(args):
"""
Calculate regulon enrichment (as AUC values) for cells.
"""
LOGGER.info("Loading expression matrix.")
ex_mtx = _load_expression_matrix(args)
if any(args.regulons_fname.name.endswith(ext) for ext in FILE_EXTENSION2SEPARATOR.keys()):
LOGGER.info("Creating regulons.")
regulons = _df2regulons(args.regulons_fname.name)
elif args.regulons_fname.name.endswith('.gmt'):
LOGGER.info("Loading regulons.")
regulons = GeneSignature.from_gmt(args.regulons_fname.name,
field_separator='\t', gene_separator='\t')
else:
LOGGER.info("Loading regulons.")
regulons = _load_modules(args.regulons_fname.name)
LOGGER.info("Calculating enrichment.")
auc_heatmap = aucell(ex_mtx, regulons, auc_threshold=args.auc_threshold,
noweights=args.weights != 'yes', num_workers=args.num_workers)
LOGGER.info("Writing results to file.")
auc_heatmap.to_csv(args.output)
def test_add():
gss = GeneSignature.from_gmt(field_separator='\t',
gene_separator='\t',
fname=TEST_SIGNATURE_FNAME)
res = gss[0].add("MEF2")
assert "MEF2" in res
def gs():
return GeneSignature.from_gmt(TEST_SIGNATURE_FNAME,
gene_separator="\t", field_separator="\t", )[0]
