def compress_vcf(path, data_path, file_name):
data_bgz = None
with open(path, 'rb') as data_stream:
data_bgz = "{}/{}.bgz".format(data_path, file_name)
with open(data_bgz, 'wb') as bgz_stream:
with bgzip.BGZipWriter(bgz_stream) as fh:
fh.writelines(data_stream.readlines())
return data_bgz
def write_body_in_gz_file(self):
""" Writes body in the compressed file, or on the stdout, regarding input arguments. """
if self.path:
with open(self.path, "a+b") as raw:
with bgzip.BGZipWriter(raw) as self.file:
for list_item in self.list_of_body_records_chrom:
self.file.write(list_item.line.encode('utf-8'))
else:
for list_item in self.list_of_body_records_chrom:
print(list_item.line.encode('utf-8'))
def created_vcf_tabix_files(vcf_content):
_, tmp_input_path = tempfile.mkstemp(prefix='vw_test_file_', suffix='.vcf.gz')
with open(tmp_input_path, 'wb') as raw_fd:
with bgzip.BGZipWriter(raw_fd) as fh:
fh.write(vcf_content)
tabix_cmd_response = subprocess.run(['tabix', '-p', 'vcf', raw_fd.name])
tabix_cmd_response.check_returncode()
files_path = (raw_fd.name, raw_fd.name + ".tbi")
created_files.append(files_path)
return files_path
def write_header_in_gz_file(self):
""" Writes header in the compressed file, or on the stdout, regarding input arguments. """
if self.path:
with open(self.path, "w+b") as raw:
with bgzip.BGZipWriter(raw) as self.file:
self.file.write(self.version.encode('utf-8'))
for list_item in self.list_of_header_objects:
self.file.write(list_item.line.encode('utf-8'))
self.file.write(self.body_header_line.line.encode('utf-8'))
else:
for list_item in self.list_of_header_objects:
print(list_item.line.encode('utf-8'))
def partialize_vcf(uri: str, number_of_lines: int, zip_format: str="bgzip") -> bytes:
with io.BytesIO() as raw:
if "bgzip" == zip_format:
zip_writer = bgzip.BGZipWriter(raw)
elif "gzip" == zip_format:
zip_writer = gzip.GzipFile(fileobj=raw, mode="w")
else:
raise ValueError("Supported values for `zip_format` are 'bgzip' and 'gzip'")
with zip_writer as writer:
for line in _vcf_lines(uri, number_of_lines):
writer.write(line)
out = raw.getvalue()
return out
def test_profile_write(self):
print()
with open("tests/fixtures/partial.vcf.gz", "rb") as raw:
with gzip.GzipFile(fileobj=raw) as fh:
inflated_data = fh.read()
with profile("gzip write"):
with gzip.GzipFile(fileobj=io.BytesIO(), mode="w") as fh:
fh.write(inflated_data)
for num_threads in range(1, 1 + bgzip.available_cores):
with profile(f"bgzip write (num_threads={num_threads})"):
with bgzip.BGZipWriter(io.BytesIO(),
num_threads=num_threads) as writer:
n = 987345
writer.write(inflated_data[:n])
writer.write(inflated_data[n:])
print()
def install_cache_manual_fasta(vep_version, assembly, cache_path,
homo_sapiens_path, index_path):
dest_dir = "{}/{}_{}".format(homo_sapiens_path, vep_version, assembly)
fasta_names = {
"GRCh37": "Homo_sapiens.GRCh37.75.dna."
"primary_assembly.fa.gz",
"GRCh38": "Homo_sapiens.GRCh38.dna.toplevel.fa.gz"
}
fasta_name = fasta_names[assembly]
fasta_type = {"GRCh37": "dna", "GRCh38": "dna_index"}
download_file(
"ftp://ftp.ensembl.org/pub/release-{vep_ver}"
"/fasta/homo_sapiens/{type}/"
"{fasta_name}".format(
vep_ver=75 if assembly == 'GRCh37' else vep_version,
type=fasta_type[assembly],
fasta_name=fasta_name),
"{}/{fasta_name}".format(dest_dir, fasta_name=fasta_name), 1800)
fasta_path = "{}/{}".format(dest_dir, fasta_name)
file_info = magic.from_file(fasta_path)
# convert gzip to bgzip if necessary for indexing
# bgzipped files are described as "gzip file with extra field"
if "extra field" not in file_info:
contents = None
with gzip.open(fasta_path, 'r') as gzip_stream:
contents = gzip_stream.readlines()
with open(fasta_path, 'wb') as bgzip_stream:
with bgzip.BGZipWriter(bgzip_stream) as fh:
fh.writelines(contents)
get_fai_and_gzi = "samtools faidx {dest}/{fasta_name}" \
.format(dest=dest_dir, fasta_name=fasta_names[assembly])
results = subprocess.run(get_fai_and_gzi.split())
if results.returncode == 0:
open(index_path, 'w').close()
else:
raise ReferenceDownloadError("Error creating index. Please try again.")
def test_pathalogical_write(self):
fh = io.BytesIO()
with bgzip.BGZipWriter(fh):
fh.write(b"")