def f():
snps = SNPs("tests/input/generic.csv")
self.assertEqual(
os.path.relpath(snps.save_snps(sep=",")),
f"output{os.sep}generic_GRCh37.csv",
)
self.run_parsing_tests("output/generic_GRCh37.csv", "generic")
def test_save_snps_vcf_phased(self):
# read phased data
s = SNPs("tests/input/testvcf_phased.vcf")
# setup resource to use test FASTA reference sequence
r = Resources()
r._reference_sequences["GRCh37"] = {}
with open("tests/input/generic.fa", "rb") as f_in:
with atomic_write("tests/input/generic.fa.gz",
mode="wb",
overwrite=True) as f_out:
with gzip.open(f_out, "wb") as f_gzip:
shutil.copyfileobj(f_in, f_gzip)
seq = ReferenceSequence(ID="1", path="tests/input/generic.fa.gz")
r._reference_sequences["GRCh37"]["1"] = seq
# save phased data to VCF
assert os.path.relpath(
s.save_snps(vcf=True)) == "output/vcf_GRCh37.vcf"
# read saved VCF
s = SNPs("output/vcf_GRCh37.vcf")
assert s.phased
pd.testing.assert_frame_equal(s.snps, self.generic_snps_vcf())
def f():
with tempfile.TemporaryDirectory() as tmpdir:
snps = SNPs("tests/input/generic.csv", output_dir=tmpdir)
dest = os.path.join(tmpdir, "generic_GRCh37.csv")
self.assertEqual(
snps.save_snps(sep=","),
dest,
)
self.run_parsing_tests(dest, "generic")
def test_save_snps_csv_phased(self):
# read phased data
s = SNPs("tests/input/testvcf_phased.vcf")
# save phased data to CSV
assert os.path.relpath(s.save_snps()) == "output/vcf_GRCh37.csv"
# read saved CSV
s = SNPs("output/vcf_GRCh37.csv")
assert s.phased
pd.testing.assert_frame_equal(s.snps, self.generic_snps_vcf())
"""
Validate input VCF files & remap them to GRCh37.
depends on:
> python 3
> argparse==1.4.0
> snps==0.4.0
> io
"""
import argparse
from snps import SNPs
import io
parser = argparse.ArgumentParser(description='Remap VCF files to GRCh37')
parser.add_argument('-i', '--input_file', help='Input VCF file')
parser.add_argument('-o', '--output_file', help='Output VCF file basename')
args = vars(parser.parse_args())
input_file = args['input_file']
output_file = args['output_file']
output_file_name = f"{output_file}.vcf"
# read & validate input file
snps = SNPs(input_file)
# remap SNPs if reference genome is not GRCh37
if snps.build_detected and snps.build != 37:
snps.remap_snps(37)
# save to file
saved_snps = snps.save_snps(output_file_name, sep="\t", header=False, vcf=True)
)
parser.add_argument(
'-t',
'--input_target',
help=
'Input BIM file (a combination of all BIM files, transformed into a 23andme-like format'
)
parser.add_argument(
'-b',
'--input_base',
help='Input base file, transformed into a 23andme-like format')
args = vars(parser.parse_args())
# Args to variable
input_target = args['input_target']
input_base = args['input_base']
###############################################################################
# Detect builds and update the base's build if it does not match the target's #
###############################################################################
target = SNPs(input_target, output_dir='.')
base = SNPs(input_base, output_dir='.')
if base.build != target.build:
base.remap_snps(target.build)
updated_base = base.save_snps("new_base_coordinates.txt",
sep="\t",
header=True)
class TestSnps(BaseSNPsTestCase):
def setUp(self):
self.snps_GRCh38 = SNPs("tests/input/GRCh38.csv")
self.snps = SNPs("tests/input/chromosomes.csv")
self.snps_only_detect_source = SNPs("tests/input/chromosomes.csv",
only_detect_source=True)
self.snps_none = SNPs(None)
with open("tests/input/chromosomes.csv", "r") as f:
self.snps_buffer = SNPs(f.read().encode("utf-8"))
with atomic_write("tests/input/chromosomes.csv.zip",
mode="wb",
overwrite=True) as f:
with zipfile.ZipFile(f, "w") as f_zip:
f_zip.write("tests/input/chromosomes.csv",
arcname="chromosomes.csv")
with open("tests/input/chromosomes.csv.zip", "rb") as f:
data = f.read()
self.snps_buffer_zip = SNPs(data)
os.remove("tests/input/chromosomes.csv.zip")
with open("tests/input/chromosomes.csv", "rb") as f_in:
with atomic_write("tests/input/chromosomes.csv.gz",
mode="wb",
overwrite=True) as f_out:
with gzip.open(f_out, "wb") as f_gzip:
shutil.copyfileobj(f_in, f_gzip)
with open("tests/input/chromosomes.csv.gz", "rb") as f:
data = f.read()
self.snps_buffer_gz = SNPs(data)
os.remove("tests/input/chromosomes.csv.gz")
def snps_discrepant_pos(self):
return self.create_snp_df(rsid=["rs3094315"],
chrom=["1"],
pos=[1],
genotype=["AA"])
def test_assembly(self):
assert self.snps_GRCh38.assembly == "GRCh38"
def test_assembly_no_snps(self):
assert self.snps_none.assembly == ""
def test_snp_buffer_zip(self):
assert self.snps_buffer_zip.snp_count == 6
def test_snp_buffer_gz(self):
assert self.snps_buffer_gz.snp_count == 6
def test_snp_buffer(self):
assert self.snps_buffer.snp_count == 6
def test_snp_count(self):
assert self.snps.snp_count == 6
def test_snp_count_no_snps(self):
assert self.snps_none.snp_count == 0
def test_chromosomes(self):
assert self.snps.chromosomes == ["1", "2", "3", "5", "PAR", "MT"]
def test_chromosomes_no_snps(self):
assert self.snps_none.chromosomes == []
def test_chromosomes_summary(self):
assert self.snps.chromosomes_summary == "1-3, 5, PAR, MT"
def test_chromosomes_summary_no_snps(self):
assert self.snps_none.chromosomes_summary == ""
def test_build_no_snps(self):
assert not self.snps_none.build
def test_build_detected_no_snps(self):
assert not self.snps_none.build_detected
def test_build_detected_PAR_snps(self):
if (not os.getenv("DOWNLOADS_ENABLED")
or os.getenv("DOWNLOADS_ENABLED") == "true"):
snps = SNPs("tests/input/GRCh37_PAR.csv")
assert snps.build == 37
assert snps.build_detected
def test_sex_no_snps(self):
assert self.snps_none.sex == ""
def test_sex_Male_Y_chrom(self):
s = self.simulate_snps(chrom="Y",
pos_start=1,
pos_max=59373566,
pos_step=10000)
file = s.save_snps()
snps = SNPs(file)
assert snps.sex == "Male"
def test_get_summary(self):
assert self.snps_GRCh38.get_summary() == {
"source": "generic",
"assembly": "GRCh38",
"build": 38,
"build_detected": True,
"snp_count": 4,
"chromosomes": "1, 3",
"sex": "",
}
def test_get_summary_no_snps(self):
assert not self.snps_none.get_summary()
def test_is_valid_True(self):
assert self.snps_GRCh38.is_valid()
def test_is_valid_False(self):
assert not self.snps_none.is_valid()
def test__read_raw_data(self):
assert self.snps_none.snps.empty
assert self.snps_none.source == ""
def test__lookup_build_with_snp_pos_None(self):
snps = SNPs()
snps._snps = self.snps_discrepant_pos()
assert not snps.detect_build()
def test_get_assembly_None(self):
snps = SNPs()
assert snps.get_assembly() is ""
def test_save_snps_source(self):
assert (os.path.relpath(
self.snps_GRCh38.save_snps()) == "output/generic_GRCh38.csv")
snps = SNPs("output/generic_GRCh38.csv")
pd.testing.assert_frame_equal(snps.snps, self.snps_GRCh38.snps)
def test_save_snps_buffer(self):
out = io.StringIO()
self.snps.save_snps(out)
assert out.read().startswith("# Generated by snps")
def test_snps_only_detect_source(self):
assert self.snps_only_detect_source.source == "generic"
def test_duplicate_rsids(self):
snps = SNPs("tests/input/duplicate_rsids.csv")
result = self.create_snp_df(rsid=["rs1"],
chrom=["1"],
pos=[101],
genotype=["AA"])
duplicate_snps = self.create_snp_df(rsid=["rs1", "rs1"],
chrom=["1", "1"],
pos=[102, 103],
genotype=["CC", "GG"])
pd.testing.assert_frame_equal(snps.snps, result)
pd.testing.assert_frame_equal(snps.duplicate_snps, duplicate_snps)
def test_deduplicate_false(self):
snps = SNPs("tests/input/duplicate_rsids.csv", deduplicate=False)
result = self.create_snp_df(
rsid=["rs1", "rs1", "rs1"],
chrom=["1", "1", "1"],
pos=[101, 102, 103],
genotype=["AA", "CC", "GG"],
)
pd.testing.assert_frame_equal(snps.snps, result)
def test_save_snps_no_snps(self):
s = SNPs()
assert not s.save_snps()
def test_save_snps_specify_file(self):
s = SNPs("tests/input/GRCh37.csv")
assert os.path.relpath(s.save_snps("snps.csv")) == "output/snps.csv"
s_saved = SNPs("output/snps.csv")
pd.testing.assert_frame_equal(s_saved.snps, self.snps_GRCh37())
def test_save_snps(self):
snps = SNPs("tests/input/GRCh37.csv")
assert os.path.relpath(snps.save_snps()) == "output/generic_GRCh37.csv"
s_saved = SNPs("output/generic_GRCh37.csv")
pd.testing.assert_frame_equal(s_saved.snps, self.snps_GRCh37())
