def test_omit_bad_seqs(self):
"""correctly omit bad sequences from an alignment"""
data = {
"s1": "---ACC---TT-",
"s2": "---ACC---TT-",
"s3": "---ACC---TT-",
"s4": "--AACCG-GTT-",
"s5": "--AACCGGGTTT",
"s6": "AGAACCGGGTT-",
"s7": "------------",
}
aln = make_aligned_seqs(data=data, moltype=DNA)
# default just eliminates strict gap sequences
dropbad = sample.omit_bad_seqs()
got = dropbad(aln)
expect = data.copy()
del expect["s7"]
self.assertEqual(got.to_dict(), expect)
# providing a more stringent gap_frac
dropbad = sample.omit_bad_seqs(gap_fraction=0.5)
got = dropbad(aln)
expect = data.copy()
for n in ("s1", "s2", "s3", "s7"):
del expect[n]
self.assertEqual(got.to_dict(), expect)
# setting quantile drops additional sequences
dropbad = sample.omit_bad_seqs(quantile=6 / 7)
got = dropbad(aln)
expect = data.copy()
for n in ("s6", "s7"):
del expect[n]
self.assertEqual(got.to_dict(), expect)
def _get_all_composables(tmp_dir_name):
test_model1 = evo.model("HKY85")
test_model2 = evo.model("GN")
test_hyp = evo.hypothesis(test_model1, test_model2)
test_num_reps = 100
applications = [
align.align_to_ref(),
align.progressive_align(model="GY94"),
evo.ancestral_states(),
evo.bootstrap(hyp=test_hyp, num_reps=test_num_reps),
evo.hypothesis(test_model1, test_model2),
evo.model("GN"),
evo.tabulate_stats(),
sample.fixed_length(100),
sample.min_length(100),
io.write_db(tmp_dir_name, create=True),
io.write_json(tmp_dir_name, create=True),
io.write_seqs(tmp_dir_name, create=True),
sample.omit_bad_seqs(),
sample.omit_degenerates(),
sample.omit_duplicated(),
sample.take_codon_positions(1),
sample.take_named_seqs(),
sample.trim_stop_codons(gc=1),
translate.select_translatable(),
tree.quick_tree(),
tree.scale_branches(),
tree.uniformize_tree(),
]
return applications
def _get_all_composable_apps():
applications = [
align.align_to_ref(),
align.progressive_align(model="GY94"),
sample.fixed_length(100),
sample.min_length(100),
io.write_seqs(os.getcwd()),
sample.omit_bad_seqs(),
sample.omit_degenerates(),
sample.take_codon_positions(1),
sample.take_named_seqs(),
sample.trim_stop_codons(gc=1),
]
return applications