def test_convert_counts(tmp_path, counts, expfile):
csvfile = str(tmp_path / "out.csv")
result = TypingResult(fromfile=data_file("prof/deep-filt-clean.json"))
result.dump_csv(csvfile, "MySample", counts=counts)
observed = pd.read_csv(csvfile)
expected = pd.read_csv(data_file(expfile))
assert observed.equals(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_filter_dupl_marker():
config = pd.read_csv(data_file("filters-redundant.csv"),
sep=None,
engine="python")
result = TypingResult(fromfile=data_file("prof/deep-raw.json"))
message = "filter config file contains duplicate entries for some markers"
with pytest.raises(ValueError, match=message):
result.filter(static=5, dynamic=0.02, config=config)
def test_typing_rate():
result = TypingResult(fromfile=data_file("prof/two-contrib-even.json"))
rates = result.typing_rate()
assert rates.TypedReads.head(5).to_list() == [3427, 2653, 3241, 4105, 3819]
assert rates.TotalReads.head(5).to_list() == [4550, 4540, 4539, 4531, 4538]
expected = [0.753187, 0.584361, 0.714034, 0.905981, 0.841560]
observed = rates.TypingRate.head(5).to_list()
for exp, obs in zip(expected, observed):
assert exp == pytest.approx(obs)
def test_filter_missing_column():
config = pd.read_csv(data_file("filters-missing.csv"),
sep=None,
engine="python")
result = TypingResult(fromfile=data_file("prof/deep-raw.json"))
with pytest.raises(
ValueError,
match=r"filter config file missing column\(s\): Static"):
result.filter(static=5, dynamic=0.02, config=config)
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_type_cli_simple(tmp_path):
outfile = str(tmp_path / "typing-result.json")
arglist = [
"type",
"--out",
outfile,
data_file("pashtun-sim/tiny-panel.tsv"),
data_file("pashtun-sim/aligned-reads.bam"),
]
args = microhapulator.cli.get_parser().parse_args(arglist)
microhapulator.cli.type.main(args)
result = TypingResult(fromfile=outfile)
assert result.haplotypes("mh13KK-218") == set()
assert result.data["markers"]["mh13KK-218"]["typing_result"] == {
"C,T,C,G": 1,
"C,T,T,T": 1,
"G,T,C,T": 1,
"T,A,C,T": 1,
"T,G,C,T": 3,
"T,G,T,T": 2,
"T,T,A,T": 5,
"T,T,C,A": 1,
"T,T,C,C": 2,
"T,T,C,G": 2,
"T,T,C,T": 1178,
"T,T,G,T": 2,
"T,T,T,A": 2,
"T,T,T,G": 6,
"T,T,T,T": 1170,
}
assert result.haplotypes("mh21KK-320") == set()
assert result.data["markers"]["mh21KK-320"]["typing_result"] == {
"G,A,A,A": 1,
"G,A,C,A": 3,
"G,A,G,A": 3,
"G,A,T,A": 1075,
"G,A,T,C": 1,
"G,A,T,G": 1,
"G,A,T,T": 2,
"G,C,C,A": 1,
"G,C,T,A": 4,
"G,G,A,A": 2,
"G,G,A,T": 1,
"G,G,C,A": 1075,
"G,G,C,C": 3,
"G,G,C,G": 12,
"G,G,C,T": 5,
"G,G,T,A": 4,
"G,T,C,A": 1,
"T,G,C,A": 1,
}
def test_filter_simple():
bam = data_file("pashtun-sim/aligned-reads.bam")
tsv = data_file("pashtun-sim/tiny-panel.tsv")
observed = mhapi.type(bam, tsv)
observed.filter(static=10, dynamic=0.05)
expected = TypingResult(fromfile=data_file("pashtun-sim/test-output.json"))
assert observed == expected
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_filter_cli_config(tmp_path):
unfiltered = data_file("prof/deep-raw.json")
filtered = str(tmp_path / "genotype-call.json")
arglist = [
"filter",
unfiltered,
"--out",
filtered,
"--static",
"5",
"--dynamic",
"0.02",
"--config",
data_file("filters.csv"),
]
args = microhapulator.cli.get_parser().parse_args(arglist)
microhapulator.cli.filter.main(args)
observed = TypingResult(fromfile=filtered)
expected = TypingResult(fromfile=data_file("prof/deep-filt.json"))
assert observed == expected
def test_filter_cli(tmp_path):
unfiltered = str(tmp_path / "typing-result.json")
filtered = str(tmp_path / "genotype-call.json")
arglist = [
"type",
"--out",
unfiltered,
data_file("pashtun-sim/tiny-panel.tsv"),
data_file("pashtun-sim/aligned-reads.bam"),
]
args = microhapulator.cli.get_parser().parse_args(arglist)
microhapulator.cli.type.main(args)
arglist = [
"filter", "--out", filtered, "--static", "5", "--dynamic", "0.05",
unfiltered
]
args = microhapulator.cli.get_parser().parse_args(arglist)
microhapulator.cli.filter.main(args)
observed = TypingResult(fromfile=filtered)
expected = TypingResult(fromfile=data_file("pashtun-sim/test-output.json"))
assert observed == expected
def main(args):
result = TypingResult(fromfile=args.result)
config = None
if args.config:
config = pd.read_csv(args.config, sep=None, engine="python")
result.filter(static=args.static, dynamic=args.dynamic, config=config)
result.dump(args.out)
def type(bamfile, markertsv, minbasequal=10, max_depth=1e6):
"""Perform haplotype calling
:param str bamfile: path of a BAM file containing NGS reads aligned to marker reference sequences and sorted
:param str markertsv: path of a TSV file containing marker metadata, specifically the offset of each SNP for every marker in the panel
:param int minbasequal: minimum base quality (PHRED score) to be considered reliable for haplotype calling; default is 10, corresponding to Q10, i.e., 90% probability that the base call is correct
:param float max_depth: maximum permitted read depth
:returns: an unfiltered catalog of haplotype counts for each marker (a *typing result*)
:rtype: microhapulator.profile.TypingResult
"""
check_index(bamfile)
bam = pysam.AlignmentFile(bamfile, "rb")
markers = load_marker_definitions(markertsv)
offsets = defaultdict(list)
for n, row in markers.iterrows():
offsets[row.Marker].append(row.Offset)
cross_check_marker_ids(bam.references, offsets.keys(), "read alignments",
"marker definitions")
haplotype_caller = tally_haplotypes(bam,
offsets,
minbasequal=minbasequal,
max_depth=max_depth)
result = TypingResult()
for locusid, cov_by_pos, htcounts, ndiscarded in haplotype_caller:
result.record_coverage(locusid, cov_by_pos, ndiscarded=ndiscarded)
for haplotype, count in htcounts.items():
result.record_haplotype(locusid, haplotype, count)
return result
def test_filter_config_file():
config = pd.read_csv(data_file("filters.csv"), sep=None, engine="python")
result = TypingResult(fromfile=data_file("prof/deep-raw.json"))
result.filter(static=5, dynamic=0.02, config=config)
assert len(result.haplotypes("mh01XYZ-1")) == 8
assert len(result.haplotypes("mh02XYZ-2")) == 2
assert len(result.haplotypes("mh02XYZ-3")) == 2
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 main(args):
result = TypingResult(fromfile=args.input)
chisq, data = mhapi.interlocus_balance(
result,
include_discarded=args.discarded,
terminal=not args.quiet,
tofile=args.figure,
title=args.title,
figsize=args.figsize,
dpi=args.dpi,
)
print(f"Extent of imbalance (chi-square statistic): {chisq:.4f}")
if args.csv:
data.to_csv(args.csv, index=False)
def main(args):
result = TypingResult(fromfile=args.input)
tstat, data = mhapi.heterozygote_balance(
result,
tofile=args.figure,
title=args.title,
figsize=args.figsize,
dpi=args.dpi,
dolabels=args.labels,
absolute=args.absolute,
)
print(f"Extent of imbalance (t-statistic): {tstat:.4f}")
if args.csv:
data.to_csv(args.csv, index=False)
def main(args):
result = TypingResult(fromfile=args.result)
result.dump_csv(args.out,
args.sample,
counts=args.counts,
fix_homo=args.fix_homo)
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_dist_gujarati(gt1, gt2, dist):
r1 = TypingResult(data_file(gt1))
r2 = TypingResult(data_file(gt2))
assert mhapi.dist(r1, r2) == dist
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()