def test_mixture_failure_modes():
freqs = microhapulator.load_marker_frequencies(
data_file("freq/russ4-freq.tsv"))
markers = microhapulator.load_marker_definitions(
data_file("def/russ4-offsets.tsv"))
seqs = microhapulator.load_marker_reference_sequences(
data_file("refr/russ4-refr.fasta.gz"))
profiles = [mhapi.sim(freqs) for _ in range(3)]
with pytest.raises(ValueError,
match=r"number of profiles must match number of seeds"):
for read in mhapi.seq(profiles, markers, seqs, seeds=[42, 1776]):
pass
with pytest.raises(
ValueError,
match=r"mismatch between contributor number and proportions"):
for read in mhapi.seq(profiles,
markers,
seqs,
proportions=[0.5, 0.3, 0.1, 0.1]):
pass
message = r"specified proportions result in 0 reads for 1 or more individuals"
with pytest.raises(ValueError, match=message):
for read in mhapi.seq(profiles,
markers,
seqs,
totalreads=500,
proportions=[1, 100, 10000]):
pass
def test_uneven_mixture(capsys):
freqs = microhapulator.load_marker_frequencies(
data_file("freq/russ4-freq.tsv"))
markers = microhapulator.load_marker_definitions(
data_file("def/russ4-offsets.tsv"))
seqs = microhapulator.load_marker_reference_sequences(
data_file("refr/russ4-refr.fasta.gz"))
profiles = [mhapi.sim(freqs) for _ in range(3)]
sequencer = mhapi.seq(profiles,
markers,
seqs,
totalreads=500,
proportions=[0.5, 0.3, 0.2])
for read in sequencer:
pass
terminal = capsys.readouterr()
assert "numreads=250" in terminal.err
assert "numreads=150" in terminal.err
assert "numreads=100" in terminal.err
def test_even_mixture():
seed = numpy.random.randint(1, 2**32 - 1)
print("Seed:", seed)
numpy.random.seed(seed)
freqs = microhapulator.load_marker_frequencies(
data_file("freq/acb-dozen-freq.tsv"))
markers = microhapulator.load_marker_definitions(
data_file("def/acb-dozen-offsets.tsv"))
seqs = microhapulator.load_marker_reference_sequences(
data_file("refr/acb-dozen-refr.fasta"))
profiles = list()
for _ in range(numpy.random.randint(2, 6)):
p = mhapi.sim(freqs)
profiles.append(p)
sequencer = mhapi.seq(profiles, markers, seqs, totalreads=1000)
for n, read1, read2 in sequencer:
pass
numfragments = n * 2
assert numfragments == pytest.approx(1000, abs=50)
def test_load_marker_frequencies_missing_or_bad_columns(tsv):
message = r"column\(s\) missing from marker frequency file: Haplotype"
with pytest.raises(ValueError, match=message):
microhapulator.load_marker_frequencies(data_file(tsv))
def test_load_marker_frequencies_extra_column_ok():
freqs = microhapulator.load_marker_frequencies(
data_file("freq/korea-5loc-freq-extracol.tsv"))
assert list(freqs.columns) == ["Marker", "Haplotype", "Frequency", "Foo"]
assert freqs.Haplotype.iloc[1] == "C,A,G,G"
def test_load_marker_frequencies(tsv, nrows, value):
freqs = microhapulator.load_marker_frequencies(data_file(tsv))
assert list(freqs.columns) == ["Marker", "Haplotype", "Frequency"]
assert freqs.shape[0] == nrows
assert freqs.Frequency.iloc[23] == pytest.approx(value)