def test_roundtrip(basictrees):
"""Make sure scaling by x and then 1/x changes nothing."""
heights = Height().consume(basictrees)
scaled_heights = build_pipeline("scale -s 2.0 | scale -s 0.5 | height",
basictrees)
for x, y in zip(heights, scaled_heights):
assert x == y
lengths = Length().consume(basictrees)
scaled_lengths = build_pipeline("scale -s 2.0 | scale -s 0.5 | length",
basictrees)
for x, y in zip(lengths, scaled_lengths):
assert x == y
def test_annotation_subtree(basictrees):
subtrees = build_pipeline(
"annotate -f tests/argfiles/annotation.csv -k taxon | subtree --attribute f1 --values 0",
basictrees)
expected_taxa = (3, 3, 3, 3, 3, 6)
for t, n in zip(subtrees, expected_taxa):
assert len(t.get_leaves()) == n
subtrees = build_pipeline(
"annotate -f tests/argfiles/annotation.csv -k taxon | subtree --attribute f1 --values 1",
basictrees)
taxa = extracted_taxa(subtrees)
subtrees = build_pipeline(
"annotate -f tests/argfiles/annotation.csv -k taxon | subtree D E F",
basictrees)
assert taxa == extracted_taxa(subtrees)
def test_categorical_annotation(treefilenewick):
# This is just to make sure the clade probability calculator doesnt't
# erroneously try to calculate means etc. of categorical annotations
list(
build_pipeline(
"annotate -f tests/argfiles/categorical_annotation.csv -k taxon | clades",
treefilenewick('basic.trees')))
def test_plot_annotated(tmpdir, basictrees, with_ete3):
annotated_trees = build_pipeline(
"annotate --f tests/argfiles/annotation.csv -k taxon",
source=basictrees)
plot = Plot(output=str(tmpdir.join('test')),
attribute="f1",
dummy=not with_ete3)
list(plot.consume(annotated_trees))
def test_implicit_source(treefilepath, argfilepath):
output = build_pipeline(
"cat -s 2 | rename -f {0} --from old --to new | prune X B | dedupe | uniq | support --sort | stat"
.format(argfilepath('rename.txt')),
source=treefilepath("basic.trees"))
for t in output:
leaves = t.get_leaf_names()
assert all((leaves.count(x) == 1 for x in leaves))
assert "A" not in leaves
assert "X" not in leaves
assert "B" not in leaves
assert all((x in leaves for x in ("C", "D", "E", "F")))
def test_pipeline(basictrees, argfilepath):
"""Silly long pipeline to stress test build_pipeline."""
output = build_pipeline(
"cat -s 2 | rename -f {0} --from old --to new | prune X B | dedupe | uniq | support --sort | stat"
.format(argfilepath('rename.txt')),
source=basictrees)
for t in output:
leaves = t.get_leaf_names()
assert all((leaves.count(x) == 1 for x in leaves))
assert "A" not in leaves
assert "X" not in leaves
assert "B" not in leaves
assert all((x in leaves for x in ("C", "D", "E", "F")))
def test_extract_annotations(treefile, argfilepath):
trees = list(NewickParser().consume(treefile('basic.trees')))
with tempfile.NamedTemporaryFile(mode="r") as fp:
list(
build_pipeline(
"annotate -f {0} -k taxon | annotate --extract -f {1}".format(
argfilepath('annotation.csv'), fp.name), trees))
fp.seek(0)
reader = csv.DictReader(fp)
assert all(
(field in reader.fieldnames for field in ("f1", "f2", "f3")))
assert "tree_number" not in reader.fieldnames
for row in reader:
if row["name"] == "A":
assert row["f1"] == "0"
assert row["f2"] == "1"
assert row["f3"] == "1"
def test_pretty_compressed_mid_signal(treefilenewick):
for x in build_pipeline("support | pretty --compress",
treefilenewick('mid_signal.trees')):
assert type(x) == str
def test_non_leaf_sibling(basictrees):
siblings = list(build_pipeline("sibling C", source=basictrees))
assert siblings == ["(A,B)", "A", "(A,B)", "(A,B)", "A", "E"]
def test_attribute_singleton_collapse(treefilenewick, argfilepath):
collapsed = build_pipeline(
"annotate -f {0} -k taxon | collapse --attribute f3".format(argfilepath('annotation.csv')),
treefilenewick('basic.trees'))
for t in collapsed:
assert len(t.get_leaves()) == 6
