def test_to_dict(self, lineage_data):
lineage1 = Lineage(**lineage_data["good"])
assert lineage1.to_dict("Class") == {
"Kingdom": "Bacteria",
"Phylum": "P1",
"Class": "C1",
}
def test_to_str(self, lineage_data):
lineage1 = Lineage(**lineage_data["good"])
assert (
lineage1.to_str(style="gg", level="Family")
== "k__Bacteria;p__P1;c__C1;o__O1;f__F1"
)
assert (
lineage1.to_str(style="silva", level="Order")
== "D_0__Bacteria;D_1__P1;D_2__C1;D_3__O1"
)
def test_sub(self, lineage_data):
lineage1 = Lineage(**lineage_data["good"])
lineage2 = Lineage(**{**lineage_data["good"], **{"Order": "O2"}})
assert lineage1 - lineage2 == lineage2 - lineage1
common = {
k: v
for k, v in lineage_data["good"].items()
if k in ["Kingdom", "Phylum", "Class"]
}
assert lineage1 - lineage2 == Lineage(**common)
def test_init(self, lineage_data):
good_lineage = Lineage(**lineage_data["good"])
assert good_lineage.Kingdom == lineage_data["good"]["Kingdom"]
assert good_lineage.Phylum == lineage_data["good"]["Phylum"]
assert good_lineage.Class == lineage_data["good"]["Class"]
assert good_lineage.Order == lineage_data["good"]["Order"]
assert good_lineage.Family == lineage_data["good"]["Family"]
assert good_lineage.Genus == lineage_data["good"]["Genus"]
assert good_lineage.Species == lineage_data["good"]["Species"]
with pytest.warns(RuntimeWarning):
Lineage(**lineage_data["bad"])
def test_taxid(self, lineage_data):
lineage1 = Lineage.from_str(
"k__Bacteria;p__Firmicutes;c__Clostridia;o__Clostridiales;f__Ruminococcaceae"
)
assert lineage1.taxid[1] == 216572
lineage2 = Lineage.from_str(
"k__Bacteria;p__Proteobacteria;c__Gammaproteobacteria;o__Enterobacterales;f__Enterobacteriaceae;g__Escherichia;s__coli"
)
assert lineage2.taxid[1] == 562
lineage3 = Lineage.from_str(
"k__Bacteria;p__Proteobacteria;c__Gammaproteobacteria;o__Enterobacterales;f__Enterobacteriaceae;g__Escherichia;s__dragon"
)
with pytest.warns(RuntimeWarning):
assert lineage3.taxid[1] == 561
def test_filter(self, stool_biom):
otu_inst = Otu(stool_biom)
query = "Firmicutes"
func = lambda values, id_, md: Lineage(**md).Phylum == query
md = otu_inst.obs_metadata
ind = md.index[md.Phylum == query]
otu_filtered = otu_inst.filter(func=func)
assert otu_filtered.otu_data.shape[1] == otu_inst.otu_data.shape[1]
assert otu_filtered.otu_data.shape[0] < otu_inst.otu_data.shape[0]
assert set(ind) == set(otu_filtered.otu_data.ids("observation"))
def test_partition(self, stool_biom):
otu_inst = Otu(stool_biom)
func = lambda id_, md: Lineage(**md).get_superset("Phylum")
md = otu_inst.obs_metadata
gen = otu_inst.partition(axis="observation", func=func)
partition_dict = {k.name[1]: v for k, v in gen}
assert set(partition_dict) == set(md.Phylum)
assert len(set(v.otu_data.shape[1]
for v in partition_dict.values())) == 1
assert (sum(
v.otu_data.shape[0]
for v in partition_dict.values()) == otu_inst.otu_data.shape[0])
def network_data(network_json_files):
""" Fixture that loads the network for the Network contructor """
data = {"good": [], "bad": []}
for kind in {"good", "bad"}:
for file in network_json_files[kind]:
with open(file, "r") as fid:
network_data = simplejson.load(fid)
non_meta_keys = ["nodes", "links"]
metadata = {k: v for k, v in network_data.items() if k not in non_meta_keys}
cmetadata = network_data.get("computational_metadata", {})
interaction_type = network_data.get("interaction_type", "")
interaction_threshold = cmetadata.get("interaction_threshold", np.nan)
pvalue_threshold = cmetadata.get("pvalue_threshold", np.nan)
pvalue_correction = None
directed = (
True if network_data.get("directionality", "") == "directed" else False
)
nodes = []
links = []
lineages = []
children_map = {}
for node in network_data["nodes"]:
try:
node_id = node["id"]
children_map[node_id] = node["children"]
except KeyError:
node_id = "missing"
nodes.append(node_id)
try:
lineage = Lineage.from_str(node["lineage"]).to_dict("Species")
except KeyError:
lineage = Lineage().to_dict("Species")
abundance = node.get("abundance", np.nan)
if abundance is not None:
lineages.append({**lineage, **dict(Abundance=abundance)})
else:
lineages.append(lineage)
obs_metadata = pd.DataFrame(lineages, index=nodes)
for link in network_data["links"]:
source, target = link.get("source", "missing"), link.get(
"target", "missing"
)
links.append(
(
source,
target,
{
"weight": link.get("weight", np.nan),
"pvalue": link.get("pvalue", np.nan),
},
)
)
data[kind].append(
(
nodes,
links,
metadata,
cmetadata,
obs_metadata,
children_map,
interaction_type,
interaction_threshold,
pvalue_threshold,
pvalue_correction,
directed,
)
)
return data
def test_get_superset(self, lineage_data):
lineage1 = Lineage(**lineage_data["good"])
lineage2 = Lineage(**{**lineage_data["good"], **{"Genus": "", "Species": ""}})
assert lineage1.get_superset("Family") == lineage2
def test_from_str(self, lineage_data):
lineage1 = Lineage(**lineage_data["good"])
lineage2 = Lineage.from_str(str(lineage1))
assert lineage1 == lineage2
assert str(lineage1) == str(lineage2)
def test_str(self, lineage_data):
lineage1 = Lineage(**lineage_data["good"])
assert str(lineage1) == "k__Bacteria;p__P1;c__C1;o__O1;f__F1;g__G1;s__S1"
lineage2 = Lineage(**{**lineage_data["good"], **{"Genus": "", "Species": ""}})
assert str(lineage2) == "k__Bacteria;p__P1;c__C1;o__O1;f__F1;g__;s__"
def test_name(self, lineage_data):
lineage1 = Lineage(**lineage_data["good"])
assert lineage1.name == ("Species", lineage_data["good"]["Species"])
lineage2 = Lineage(**{**lineage_data["good"], **{"Genus": "", "Species": ""}})
assert lineage2.name == ("Family", lineage_data["good"]["Family"])