def run(args):
start = timer()
Utilities.ensure_requisite_folders(args.output_prefix)
logging.info("Loading SNP annotation")
snp_key = KeyedDataSource.load_data(args.snp_annotation_file,
"varID",
"rsid_dbSNP150",
should_skip=KeyedDataSource.skip_na)
logging.info("Loading Genotype")
genotype, individual_ids = ModelTraining.load_genotype_folder(
args.input_genotype_folder, args.input_genotype_file_pattern, snp_key)
logging.info("Saving Genotype")
path_variant = args.output_prefix + ".variants.parquet"
Parquet.save_variants(path_variant, genotype, individual_ids)
path_metadata_variant = args.output_prefix + ".variants_metadata.parquet"
Parquet.save_metadata(path_metadata_variant, genotype)
logging.info("Processing Expression Phenotype")
expression_logic = Utilities.file_logic(
args.input_phenotype_folder, args.input_phenotype_expression_pattern)
for row in expression_logic.itertuples():
logging.info("Phenotype: %s", row.name)
process_phenotype(row.path, row.name, args.output_prefix)
end = timer()
logging.info("Finished in %s", str(end - start))
def get_gene_variant_list(model_db_folder, pattern):
logic = Utilities.file_logic(model_db_folder, pattern)
g = []
for i, l in enumerate(logic.itertuples()):
logging.log(9, "Opening db %d/%d: %s ", i + 1, logic.shape[0], l.name)
with sqlite3.connect(l.path) as connection:
w = pandas.read_sql("select * from weights;",
connection)[["gene", "varID", "rsid"]]
g.append(w)
g = pandas.concat(g).drop_duplicates()
return g
def run(args):
logging.info("Processing...")
Utilities.ensure_requisite_folders(args.output_prefix)
spec = Utilities.file_logic(args.input_folder, args.input_pattern)
with gzip.open(args.output_prefix + ".models.txt.gz", mode="w") as models:
models.write("gene\tmodel\tn\tpp\tps\n".encode())
with gzip.open(args.output_prefix + ".models_variants.txt.gz",
mode="w") as model_variants:
model_variants.write("gene\tmodel\tvariant\n".encode())
with gzip.open(args.output_prefix + ".model_summary.txt.gz",
mode="w") as model_summary:
model_summary.write(
"gene\tpes\tpes_se\tlog_nc\tlog10_nc\n".encode())
with gzip.open(args.output_prefix + ".variants_pip.txt.gz",
mode="w") as variant_pip:
variant_pip.write(
"gene\trank\tvariant_id\tpip\tlog10_abf\tcluster_id\n".
encode())
with gzip.open(args.output_prefix + ".clusters.txt.gz",
mode="w") as clusters:
clusters.write(
"gene\tcluster\tn_snps\tpip\taverage_r2\n".encode(
))
with gzip.open(args.output_prefix +
".cluster_correlations.txt.gz",
mode="w") as cluster_correlations:
cluster_correlations.write(
"gene\tid1\tid2\tvalue\n".encode())
for i, t in enumerate(spec.itertuples()):
logging.log(9, "Processing %s", t.name)
written = set()
with open(t.path) as dap:
p, pse, lognc, log10nc = None, None, None, None
for l in dap:
s = model_re.search(l)
if s:
ml = parse_model_line(t.name, s)
models.write(ml.encode())
vl = parse_model_line_for_variant(
t.name, s)
if vl:
for vl_ in vl:
model_variants.write(
vl_.encode())
continue
s = model_expected_size_re.search(l)
if s:
p, pse = parse_expected_size(s)
continue
s = lognc_re.search(l)
if s:
lognc, log10nc = parse_log_10_nc(s)
model_summary.write(
"{}\t{}\t{}\t{}\t{}\n".format(
t.name, p, pse, lognc,
log10nc).encode())
continue
s = variant_re.search(l)
if s:
rank, id, pip, log10_abvf, cluster_id = parse_variant_line(
s)
variant_pip.write(
"{}\t{}\t{}\t{}\t{}\t{}\n".
format(t.name, rank, id, pip,
log10_abvf,
cluster_id).encode())
continue
s = cluster_re.search(l)
if s:
id, n, pip, r2 = parse_cluster_line(
s)
clusters.write(
"{}\t{}\t{}\t{}\t{}\n".format(
t.name, id, n, pip,
r2).encode())
_id1 = int(id)
comps = s.group(
"correlation").strip().split()
for _id2 in range(
1,
len(comps) + 1):
if (_id1, _id2) in written or (
_id2, _id1) in written:
continue
comp = comps[_id2 - 1]
cluster_correlations.write(
"{}\t{}\t{}\t{}\n".format(
t.name, _id1, _id2,
comp).encode())
written.add((_id1, _id2))
logging.info("Finished")
