def test_collapse():
lines = fileinput.input("tests/treefiles/basic.trees")
trees = list(NewickParser().consume(lines))
collapsed = Collapse({"left":("A","B","C"), "right":("D","E","F")}).consume(trees)
# These groups are monophyletic in the first 5 of the 6 basic trees, so...
for n, t in enumerate(collapsed):
assert len(t.get_leaves()) == (2 if n < 5 else 6)
def test_length():
lines = fileinput.input("tests/treefiles/basic.trees")
trees = NewickParser().consume(lines)
lengths = Length().consume(trees)
for l in lengths:
assert type(l) == float
assert l >= 0.0
def test_file_collapse():
lines = fileinput.input("tests/treefiles/basic.trees")
trees = list(NewickParser().consume(lines))
collapsed = Collapse(filename="tests/argfiles/collapse.txt").consume(trees)
# These groups are monophyletic in the first 5 of the 6 basic trees, so...
for n, t in enumerate(collapsed):
assert len(t.get_leaves()) == (2 if n < 5 else 6)
def test_height():
lines = fileinput.input("tests/treefiles/basic.trees")
trees = NewickParser().consume(lines)
heights = Height().consume(trees)
for h in heights:
assert type(h) == float
assert h >= 0.0
def test_uniq():
lines = fileinput.input("tests/treefiles/basic.trees")
trees = list(NewickParser().consume(lines))
uniq = Uniq().consume(trees)
# The 6 basic trees comprise 5 unique topologies.
# This is a pretty weak test, but...
assert sum((1 for t in uniq)) == 5
def test_file_prune():
lines = fileinput.input("tests/treefiles/basic.trees")
trees = NewickParser().consume(lines)
pruned = Prune(filename="tests/argfiles/taxa_abc.txt").consume(trees)
for t in pruned:
leaves = t.get_leaf_names()
assert not any((x in leaves for x in ("A", "B", "C")))
def test_file_subtree():
lines = fileinput.input("tests/treefiles/basic.trees")
trees = NewickParser().consume(lines)
subtrees = Subtree(filename="tests/argfiles/taxa_abc.txt").consume(trees)
expected_taxa = (3, 3, 3, 3, 3, 6)
for t, n in zip(subtrees, expected_taxa):
assert len(t.get_leaves()) == n
def test_prune():
lines = fileinput.input("tests/treefiles/basic.trees")
trees = NewickParser().consume(lines)
pruned = Prune(["A"]).consume(trees)
for t in pruned:
leaves = t.get_leaf_names()
assert "A" not in leaves
assert all((x in leaves for x in ("B", "C", "D", "E", "F")))
def test_inverse_prune():
lines = fileinput.input("tests/treefiles/basic.trees")
trees = NewickParser().consume(lines)
pruned = Prune(["A", "B"], inverse=True).consume(trees)
for t in pruned:
leaves = t.get_leaf_names()
assert all((x in leaves for x in ("A", "B")))
assert not any((x in leaves for x in ("C", "D", "E", "F")))
def test_subtree():
subtree = Subtree.init_from_args("A,B,C")
lines = fileinput.input("tests/treefiles/basic.trees")
trees = NewickParser().consume(lines)
subtrees = subtree.consume(trees)
expected_taxa = (3, 3, 3, 3, 3, 6)
for t, n in zip(subtrees, expected_taxa):
assert len(t.get_leaves()) == n
def test_identity():
"""Make sure scaling with a factor of 1.0 changes nothing."""
lines = fileinput.input("tests/treefiles/basic.trees")
trees = list(NewickParser().consume(lines))
unscaled_trees = Scale(1.0).consume(trees)
for t1, t2 in zip(trees, unscaled_trees):
assert t1.write() == t2.write()
def test_annotation_prune():
lines = fileinput.input("tests/treefiles/basic.trees")
trees = NewickParser().consume(lines)
annotated = Annotate(filename="tests/argfiles/annotation.csv", key="taxon").consume(trees)
pruned = Prune(attribute="f1", value="0").consume(annotated)
for t in pruned:
leaves = t.get_leaf_names()
assert not any((x in leaves for x in ("A", "B", "C")))
def test_clades():
lines = fileinput.input("tests/treefiles/basic.trees")
trees = NewickParser().consume(lines)
supported = Support(filename="/dev/null").consume(trees)
for t in supported:
for n in t.traverse():
assert hasattr(n, "support")
assert type(n.support) == float
assert 0 <= n.support <= 1
def test_plot(dummy=False):
lines = fileinput.input("tests/treefiles/basic.trees")
trees = NewickParser().consume(lines)
with tempfile.NamedTemporaryFile() as fp:
plot = Plot(dummy=dummy, output=fp.name, height=600, width=800)
for x in plot.consume(trees):
pass
lines.close()
def test_categorical_annotation():
# This is just to make sure the clade probability calculator doesnt't
# erroneously try to calculate means etc. of categorical annotations
lines = fileinput.input("tests/treefiles/basic.trees")
trees = NewickParser().consume(lines)
for t in build_pipeline(
"annotate -f tests/argfiles/categorical_annotation.csv -k taxon | clades",
trees):
pass
def test_annotation_subtree():
lines = fileinput.input("tests/treefiles/basic.trees")
trees = NewickParser().consume(lines)
subtrees = build_pipeline(
"annotate -f tests/argfiles/annotation.csv -k taxon | subtree --attribute f1 --value 0",
trees)
expected_taxa = (3, 3, 3, 3, 3, 6)
for t, n in zip(subtrees, expected_taxa):
assert len(t.get_leaves()) == n
def test_scale():
scale_factor = 0.42
lines = fileinput.input("tests/treefiles/basic.trees")
trees = list(NewickParser().consume(lines))
old_heights = [t.get_farthest_leaf()[1] for t in trees]
scaled = Scale(scale_factor).consume(trees)
new_heights = [t.get_farthest_leaf()[1] for t in scaled]
for old, new in zip(old_heights, new_heights):
assert new == old * scale_factor
def test_rename_from_file():
lines = fileinput.input("tests/treefiles/basic.trees")
trees = NewickParser().consume(lines)
renamed = Rename(filename="tests/argfiles/rename.txt").consume(trees)
for t in renamed:
leaves = t.get_leaf_names()
assert "A" not in leaves
assert "X" in leaves
assert all((x in leaves for x in ("B", "C", "D", "E", "F")))
def test_stat():
lines = fileinput.input("tests/treefiles/basic.trees")
trees = NewickParser().consume(lines)
stat = Stat()
for t in stat.consume(trees):
pass
assert stat.tree_count == 6
assert stat.taxa_count == 6
assert stat.topology_count <= stat.tree_count
def test_annotate(treefile, argfilepath):
trees = NewickParser().consume(treefile('basic.trees'))
annotated = Annotate(filename=argfilepath("annotation.csv"),
key="taxon").consume(trees)
for t in annotated:
t.write(features=[])
for l in t.get_leaves():
assert hasattr(l, "f1")
assert hasattr(l, "f2")
assert hasattr(l, "f3")
def test_attribute_collapse():
lines = fileinput.input("tests/treefiles/basic.trees")
trees = list(NewickParser().consume(lines))
annotated = Annotate("tests/argfiles/annotation.csv", "taxon").consume(trees)
# f1 in the annotations applied above corresponds to the same left/right
# split as the other tests above
collapsed = Collapse(attribute="f1").consume(annotated)
# These groups are monophyletic in the first 5 of the 6 basic trees, so...
for n, t in enumerate(collapsed):
assert len(t.get_leaves()) == (2 if n < 5 else 6)
def test_plot_annotated(dummy=False):
lines = fileinput.input("tests/treefiles/basic.trees")
trees = NewickParser().consume(lines)
annotated_trees = build_pipeline(
"annotate --f tests/argfiles/annotation.csv -k taxon", source=trees)
with tempfile.NamedTemporaryFile() as fp:
plot = Plot(output=fp.name, attribute="f1", dummy=dummy)
for x in plot.consume(annotated_trees):
pass
lines.close()
def test_monophyletic_dedupe():
lines = fileinput.input("tests/treefiles/monophyletic_dupe_taxa.trees")
trees = list(NewickParser().consume(lines))
for t in trees:
leaves = t.get_leaf_names()
assert not all(
(leaves.count(x) == 1 for x in ("A", "B", "C", "E", "F")))
deduped = Dedupe().consume(trees)
for t in deduped:
leaves = t.get_leaf_names()
assert all((leaves.count(x) == 1 for x in ("A", "B", "C", "E", "F")))
def test_annotate():
lines = fileinput.input("tests/treefiles/basic.trees")
trees = NewickParser().consume(lines)
annotated = Annotate("tests/argfiles/annotation.csv",
"taxon").consume(trees)
for t in annotated:
t.write(features=[])
for l in t.get_leaves():
assert hasattr(l, "f1")
assert hasattr(l, "f2")
assert hasattr(l, "f3")
def test_rename_with_remove():
lines = fileinput.input("tests/treefiles/basic.trees")
trees = NewickParser().consume(lines)
renamed = Rename({
"A": "X",
"B": "Y",
"C": "Z"
}, remove=True).consume(trees)
for t in renamed:
leaves = t.get_leaf_names()
assert all((x in leaves for x in ("X", "Y", "Z")))
assert not any((x in leaves for x in ("A", "B", "C", "D", "E", "F")))
def test_dedupe():
lines = fileinput.input("tests/treefiles/duplicate_taxa.trees")
trees = list(NewickParser().consume(lines))
for t in trees:
orig_leaves = t.get_leaf_names()
assert len(orig_leaves) == 6
assert orig_leaves.count("A") == 2
assert all((orig_leaves.count(x) == 1 for x in ("B", "C", "E", "F")))
deduped = Dedupe().consume(trees)
for t in deduped:
leaves = t.get_leaf_names()
assert len(leaves) == 5
assert all((leaves.count(x) == 1 for x in ("A", "B", "C", "E", "F")))
def test_min_med_max_uniq():
lines = fileinput.input("tests/treefiles/basic.trees")
trees = list(NewickParser().consume(lines))
min_uniq = Uniq(lengths="min").consume(trees)
min_lengths = Length().consume(min_uniq)
med_uniq = Uniq(lengths="median").consume(trees)
med_lengths = Length().consume(med_uniq)
max_uniq = Uniq(lengths="max").consume(trees)
max_lengths = Length().consume(max_uniq)
for l, m, L in zip(min_lengths, med_lengths, max_lengths):
assert l <= m <= L
def test_pipeline():
"""Silly long pipeline to stress test build_pipeline."""
lines = fileinput.input("tests/treefiles/basic.trees")
trees = NewickParser().consume(lines)
output = build_pipeline(
"cat -s 2 | rename -f tests/argfiles/rename.txt | prune X,B | dedupe | uniq | support --sort | stat",
source=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_roundtrip():
"""Make sure scaling by x and then 1/x changes nothing."""
lines = fileinput.input("tests/treefiles/basic.trees")
trees = list(NewickParser().consume(lines))
heights = Height().consume(trees)
scaled_heights = build_pipeline("scale -s 2.0 | scale -s 0.5 | height",
trees)
for x, y in zip(heights, scaled_heights):
assert x == y
lengths = Length().consume(trees)
scaled_lengths = build_pipeline("scale -s 2.0 | scale -s 0.5 | length",
trees)
for x, y in zip(lengths, scaled_lengths):
assert x == y
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"