def test_bed_error():
p = SimulatedProfile()
p.add(0, "BOGUS", "A,C,C")
p.add(1, "BOGUS", "A,C,C")
markers = pd.read_csv(data_file("def/loc2-offsets.tsv"), sep="\t")
with pytest.raises(ValueError, match=r"unknown marker identifier 'BOGUS'"):
print(p.bedstr(markers))
def sim(frequencies, seed=None):
"""Simulate a diploid genotype from the specified microhaplotype frequencies
:param pandas.DataFrame frequencies: population haplotype frequencies
:param int seed: seed for random number generator
:returns: a simulated genotype profile for all markers specified in the haplotype frequencies
:rtype: microhapulator.profile.SimulatedProfile
"""
profile = SimulatedProfile(ploidy=2)
if seed is None:
seed = np.random.randint(2**32 - 1)
profile.data["metadata"] = {
"HaploSeed": seed,
}
np.random.seed(seed)
markers = sorted(frequencies.Marker.unique())
for haploindex in range(2):
for marker in markers:
haplofreqs = frequencies[frequencies.Marker == marker]
haplotypes = list(haplofreqs.Haplotype)
freqs = list(haplofreqs.Frequency)
freqs = [x / sum(freqs) for x in freqs]
sampled_haplotype = np.random.choice(haplotypes, p=freqs)
profile.add(haploindex, marker, sampled_haplotype)
message = f"simulated microhaplotype variation at {len(markers)} markers"
print("[MicroHapulator::sim]", message, file=sys.stderr)
return profile
def test_diff_basic():
gt1 = SimulatedProfile(fromfile=data_file("prof/diff-comp-1.json"))
gt2 = SimulatedProfile(fromfile=data_file("prof/diff-comp-2.json"))
diff = list(mhapi.diff(gt1, gt2))
assert diff == [
("MHDBL000140", {"C,C,A,A"}, {"C,C,T,A"}),
("MHDBL000163", {"A,A,G,A,T"}, {"C,G,A,A,T"}),
]
def test_diff_nonmatching_alleles():
p1 = SimulatedProfile(fromfile=data_file("prof/red-strict-profile.json"))
p2 = SimulatedProfile(fromfile=data_file("prof/red-relaxed-profile.json"))
diff = list(mhapi.diff(p1, p2))
print(diff)
assert diff == [
("mh07CP-004", set(), {"T,T,T,A,T", "A,A,T,A,T"}),
("mh09KK-157", set(), {"G,C,C,A,T"}),
]
def test_diff_large():
gt1 = SimulatedProfile(fromfile=data_file("prof/diff-comp-1.json"))
gt2 = SimulatedProfile(fromfile=data_file("prof/diff-comp-3.json"))
diff = list(mhapi.diff(gt1, gt2))
loci = [d[0] for d in diff]
print(diff[9], diff[17], diff[21])
assert loci == [
"MHDBL000002",
"MHDBL000003",
"MHDBL000007",
"MHDBL000013",
"MHDBL000017",
"MHDBL000018",
"MHDBL000030",
"MHDBL000036",
"MHDBL000038",
"MHDBL000047",
"MHDBL000058",
"MHDBL000061",
"MHDBL000076",
"MHDBL000079",
"MHDBL000082",
"MHDBL000085",
"MHDBL000088",
"MHDBL000101",
"MHDBL000106",
"MHDBL000108",
"MHDBL000111",
"MHDBL000112",
"MHDBL000122",
"MHDBL000124",
"MHDBL000128",
"MHDBL000129",
"MHDBL000135",
"MHDBL000136",
"MHDBL000138",
"MHDBL000140",
"MHDBL000144",
"MHDBL000152",
"MHDBL000154",
"MHDBL000163",
"MHDBL000181",
"MHDBL000183",
"MHDBL000194",
"MHDBL000210",
"MHDBL000211",
"MHDBL000212",
]
assert diff[9] == ("MHDBL000047", set(), {"T,T"})
assert diff[17] == ("MHDBL000101", {"C,C,C,T"}, {"T,C,C,C"})
assert diff[21] == ("MHDBL000112", {"G,G,A,C"}, set())
def test_main(tmp_path):
outfile = str(tmp_path / "profile.json")
arglist = [
"sim",
"--out",
outfile,
"--seed",
"1985",
data_file("freq/ceu50-freq.tsv"),
]
args = microhapulator.cli.get_parser().parse_args(arglist)
microhapulator.cli.sim.main(args)
observed = SimulatedProfile(fromfile=outfile)
expected = SimulatedProfile(fromfile=data_file("prof/bitusa-profile.json"))
assert observed == expected
def test_pipe_gbr_usc10(tmp_path):
hg38 = str(tmp_path / "hg38-placeholder.fasta")
copyfile(data_file("refr/usc10-refr.fna"), hg38)
run(["bwa", "index", hg38])
arglist = [
"pipe",
data_file("refr/usc10-refr.fna"),
data_file("def/usc10-offsets.tsv"),
data_file(""),
"gbr-usc",
"--workdir",
str(tmp_path),
"--threads",
"1",
"--hg38",
hg38,
]
args = microhapulator.cli.get_parser().parse_args(arglist)
microhapulator.cli.pipe.main(args)
expected = SimulatedProfile(fromfile=data_file("prof/gbr-usc10-sim.json"))
observed = TypingResult(fromfile=tmp_path / "analysis" / "gbr-usc" /
"gbr-usc-type.json")
diff = list(mhapi.diff(observed, expected))
assert len(diff) == 0
assert (tmp_path / "report.html").is_file()
expected = pd.read_csv(data_file("gbr-usc-summary.tsv"), sep="\t")
observed = pd.read_csv(tmp_path / "analysis" / "summary.tsv", sep="\t")
assert observed.equals(expected)
def test_mix_main():
with NamedTemporaryFile(suffix=".json.gz") as outfile:
arglist = [
"mix",
"--out",
outfile.name,
data_file("prof/green-sim-gt-1.json.gz"),
data_file("prof/green-sim-gt-2.json.gz"),
data_file("prof/green-sim-gt-3.json.gz"),
]
args = microhapulator.cli.get_parser().parse_args(arglist)
microhapulator.cli.mix.main(args)
p = SimulatedProfile(fromfile=outfile.name)
testp = SimulatedProfile(
fromfile=data_file("prof/green-sim-gt-combined.json.gz"))
assert p == testp
def test_profile_roundtrip(tmp_path):
seed = numpy.random.randint(1, 2**32 - 1)
freqs = pd.read_csv(data_file("freq/asw5-freq.tsv"), sep="\t")
profile = mhapi.sim(freqs, seed=seed)
profile.dump(tmp_path / "profile.json")
test = SimulatedProfile(fromfile=tmp_path / "profile.json")
assert profile == test
assert str(profile) == str(test)
def test_dist_even_mixture():
with microhapulator.open(data_file("murica/x-obs-genotype.json"),
"r") as fh:
p1 = TypingResult(fh)
p2 = SimulatedProfile.populate_from_bed(
data_file("murica/x-sim-genotype.bed"))
assert mhapi.dist(p1, p2) == 0
assert p1 == p2
def test_haplotypes():
simprof = SimulatedProfile.populate_from_bed(data_file("gttest.bed.gz"))
typeprof = TypingResult(fromfile=data_file("prof/gttest.json"))
assert simprof.haplotypes("BoGuSlOcUs") == set()
assert typeprof.haplotypes("BoGuSlOcUs") == set()
assert simprof.haplotypes("MHDBL000135") == set(["G,C,T", "G,T,C"])
assert typeprof.haplotypes("MHDBL000135") == set(["G,C,T", "G,T,C"])
assert simprof.haplotypes("MHDBL000135", index=0) == set(["G,C,T"])
assert simprof.haplotypes("MHDBL000135", index=1) == set(["G,T,C"])
assert typeprof.haplotypes("MHDBL000135", index=0) == set()
def test_sim_obs_profile_not_equal():
simprof1 = SimulatedProfile.populate_from_bed(
data_file("gttest-mismatch1.bed.gz"))
assert simprof1 is not None
assert simprof1 != 42
assert simprof1 != 3.14159
assert simprof1 != "A,C,C,T"
typeprof1 = TypingResult(fromfile=data_file("prof/gttest.json"))
assert simprof1 != typeprof1
assert typeprof1 != simprof1
assert typeprof1 != 1985
assert typeprof1 != 98.6
simprof2 = SimulatedProfile.populate_from_bed(
data_file("gttest-mismatch2.bed.gz"))
assert simprof1 != simprof2
assert simprof2 != typeprof1
assert typeprof1 != simprof2
typeprof2 = TypingResult(fromfile=data_file("prof/gttest-altered.json"))
assert typeprof1 != typeprof2
def test_main_haplo_seq(tmp_path):
profile = str(tmp_path / "profile.json")
hapseq = str(tmp_path / "haplo.fasta")
arglist = [
"sim",
"--seed",
"293847",
"--out",
profile,
"--haplo-seq",
hapseq,
"--sequences",
data_file("refr/orange-refr.fasta"),
"--markers",
data_file("def/orange-offsets.tsv"),
data_file("freq/asw2-freq.tsv"),
]
args = microhapulator.cli.get_parser().parse_args(arglist)
microhapulator.cli.sim.main(args)
observed = SimulatedProfile(fromfile=profile)
expected = SimulatedProfile(
fromfile=data_file("prof/orange-sim-profile.json"))
assert observed == expected
assert filecmp.cmp(hapseq, data_file("orange-haplo.fasta"))
def test_dist_log_mixture():
p1 = TypingResult(data_file("murica/y-obs-genotype.json"))
p2 = SimulatedProfile.populate_from_bed(
data_file("murica/y-sim-genotype.bed"))
assert mhapi.dist(p1, p2) == 19
assert p1 != p2
def test_merge_sim_genotypes():
prof1 = SimulatedProfile()
prof1.add(0, "mh11CP-004", "C,G,G")
prof1.add(1, "mh11CP-004", "C,G,G")
prof1.add(0, "mh05KK-123", "A,C")
prof1.add(1, "mh05KK-123", "A,T")
prof2 = SimulatedProfile()
prof2.add(0, "mh11CP-004", "C,T,A")
prof2.add(1, "mh11CP-004", "C,T,G")
prof2.add(0, "mh05KK-123", "A,T")
prof2.add(1, "mh05KK-123", "A,T")
prof3 = SimulatedProfile()
prof3.add(0, "mh11CP-004", "C,G,G")
prof3.add(1, "mh11CP-004", "T,G,G")
prof3.add(0, "mh05KK-123", "G,C")
prof3.add(1, "mh05KK-123", "G,T")
profile = SimulatedProfile.merge([prof1, prof2, prof3])
markers = pd.read_csv(data_file("def/loc2-offsets.tsv"), sep="\t")
output = profile.bedstr(markers)
print(output)
assert output == ("mh05KK-123\t121\t122\tA|A|A|A|G|G\n"
"mh05KK-123\t228\t229\tC|T|T|T|C|T\n"
"mh11CP-004\t162\t163\tC|C|C|C|C|T\n"
"mh11CP-004\t163\t164\tG|G|T|T|G|G\n"
"mh11CP-004\t187\t188\tG|G|A|G|G|G\n")
def test_sim_obs_profile_equality():
simprof = SimulatedProfile.populate_from_bed(data_file("gttest.bed.gz"))
typeprof = TypingResult(fromfile=data_file("prof/gttest.json"))
assert simprof == typeprof
assert typeprof == simprof
def test_haploindexes():
simprof = SimulatedProfile.populate_from_bed(
data_file("gttest-mismatch1.bed.gz"))
assert simprof.haploindexes() == set([0, 1])
typeprof = TypingResult(data_file("pashtun-sim/test-output.json"))
assert typeprof.haploindexes() == set()
def test_diff2():
gt1 = SimulatedProfile(fromfile=data_file("prof/euramer-sim-gt.json"))
gt2 = SimulatedProfile(fromfile=data_file("prof/euramer-inf-gt.json"))
diff = list(mhapi.diff(gt1, gt2))
assert diff == [("MHDBL000018", set(), {"T,G,C,T,A"})]
def test_meaning_of_life():
freqs = pd.read_csv(data_file("freq/ceu50-freq.tsv"), sep="\t")
observed = mhapi.sim(freqs, seed=42)
expected = SimulatedProfile(
fromfile=data_file("prof/meaning-of-life.json.gz"))
assert observed == expected
def main(args):
profiles = [SimulatedProfile(pfile) for pfile in args.profiles]
combined = SimulatedProfile.merge(profiles)
with mhopen(args.out, "w") as fh:
combined.dump(fh)