def test_angstrom(self):
domain = AntismashDomain(FeatureLocation(1, 2), "test")
domain.domain_id = "query"
domain.translation = self.aligns[domain.domain_id].replace("-", "")
sig = extract_sig.get_34_aa_signature(domain)
assert sig == "L--SFDASLFEMYLLTGGDRNMYGPTEATMCATW"
def test_subtype_recognition(self):
assert self.domain.domain_id
bio = self.domain.to_biopython()[0]
assert bio.qualifiers["aSTool"] == [TOOL]
assert bio.qualifiers["domain_id"]
new = AntismashDomain.from_biopython(bio)
assert isinstance(new, ModularDomain)
class TestConversion(unittest.TestCase):
def setUp(self):
self.protein_location = FeatureLocation(0, 1)
self.domain = AntismashDomain(FeatureLocation(1, 3, 1), locus_tag="locus",
tool="test", protein_location=self.protein_location)
self.domain.domain_id = "test1"
def test_conversion(self):
domain = self.domain
domain.asf.add("first")
domain.asf.add("second")
assert domain.tool == "test"
assert domain.created_by_antismash
assert domain.locus_tag == "locus"
bio = domain.to_biopython()
assert len(bio) == 1
assert bio[0].qualifiers["aSTool"] == ["test"]
assert bio[0].qualifiers["tool"] == ["antismash"]
new_domain = AntismashDomain.from_biopython(bio[0])
assert new_domain.asf.hits == domain.asf.hits
assert new_domain.asf.hits == ["first", "second"]
assert new_domain.tool == domain.tool == "test"
assert new_domain.created_by_antismash
assert new_domain.locus_tag == "locus"
assert new_domain.protein_location == self.protein_location
def test_linebreaks(self):
self.domain.domain_id = "some-long-name.Condensation.1"
bio = self.domain.to_biopython()
# pretend a genbank was written and is being read back in
# but the line was long enough to cause a line break
bio[0].qualifiers["domain_id"][0] = self.domain.domain_id.replace(".", ". ")
new = AntismashDomain.from_biopython(bio[0])
assert new.domain_id == self.domain.domain_id
def test_linebreaks(self):
self.domain.domain_id = "some-long-name.Condensation.1"
bio = self.domain.to_biopython()
# pretend a genbank was written and is being read back in
# but the line was long enough to cause a line break
bio[0].qualifiers["domain_id"][0] = self.domain.domain_id.replace(".", ". ")
new = AntismashDomain.from_biopython(bio[0])
assert new.domain_id == self.domain.domain_id
def generate_domains(self):
inputs = fasta.read_fasta(
path.get_full_path(__file__, 'data', 'PKS_KS.input'))
domains = []
last_end = 0
for translation in inputs.values():
location = FeatureLocation(last_end + 10,
last_end + len(translation) * 3 + 16)
domain = AntismashDomain(location, tool="test")
domain.translation = translation
domains.append(domain)
domain.domain = "PKS_KS"
location = FeatureLocation(
last_end + 10, last_end + len(domains[-1].translation) * 3 + 16)
domains.append(AntismashDomain(location, tool="test"))
domains[-1].domain = "PKS_KR"
return domains
def test_conversion(self):
protein_location = FeatureLocation(0, 1)
domain = AntismashDomain(FeatureLocation(1, 3, 1), locus_tag="locus",
tool="test", protein_location=protein_location)
domain.domain_subtype = "subtest"
domain.specificity = ["a", "c", "f"]
domain.asf.add("first")
domain.asf.add("second")
assert domain.tool == "test"
assert domain.created_by_antismash
assert domain.locus_tag == "locus"
bio = domain.to_biopython()
assert len(bio) == 1
assert bio[0].qualifiers["aSTool"] == ["test"]
assert bio[0].qualifiers["tool"] == ["antismash"]
new_domain = AntismashDomain.from_biopython(bio[0])
assert new_domain.domain_subtype == domain.domain_subtype == "subtest"
assert new_domain.specificity == domain.specificity == ["a", "c", "f"]
assert new_domain.asf.hits == domain.asf.hits
assert new_domain.asf.hits == ["first", "second"]
assert new_domain.tool == domain.tool == "test"
assert new_domain.created_by_antismash
assert new_domain.locus_tag == "locus"
assert new_domain.protein_location == protein_location
def test_conversion(self):
domain = self.domain
domain.asf.add("first")
domain.asf.add("second")
assert domain.tool == "test"
assert domain.created_by_antismash
assert domain.locus_tag == "locus"
bio = domain.to_biopython()
assert len(bio) == 1
assert bio[0].qualifiers["aSTool"] == ["test"]
assert bio[0].qualifiers["tool"] == ["antismash"]
new_domain = AntismashDomain.from_biopython(bio[0])
assert new_domain.asf.hits == domain.asf.hits
assert new_domain.asf.hits == ["first", "second"]
assert new_domain.tool == domain.tool == "test"
assert new_domain.created_by_antismash
assert new_domain.locus_tag == "locus"
assert new_domain.protein_location == self.protein_location
def setUp(self):
self.protein_location = FeatureLocation(0, 1)
self.domain = AntismashDomain(FeatureLocation(1, 3, 1),
locus_tag="locus",
tool="test",
protein_location=self.protein_location)
def test_subvariant_detection(self):
bio = self.rre.to_biopython()[0]
new = AntismashDomain.from_biopython(bio)
assert isinstance(new, RREDomain)
def test_domain_passthrough(self):
for value in ["test", "DomainName"]:
domain = AntismashDomain(FeatureLocation(1, 3, 1), locus_tag="locus",
tool="test", protein_location=FeatureLocation(0, 1),
domain=value)
assert domain.domain == value
def generate_domain_features(
gene: CDSFeature,
domains: List[HMMResult]) -> Dict[HMMResult, AntismashDomain]:
""" Generates AntismashDomain features for each provided HMMResult
Arguments:
gene: the CDSFeature the domains were found in
domains: a list of HMMResults found in the CDSFeature
Returns:
a dictionary mapping the HMMResult used to the matching AntismashDomain
"""
new_features = {}
domain_counts = defaultdict(int) # type: Dict[str, int]
for domain in domains:
loc = gene.get_sub_location_from_protein_coordinates(
domain.query_start, domain.query_end)
prot_loc = FeatureLocation(domain.query_start, domain.query_end)
# set up new feature
new_feature = AntismashDomain(loc,
tool="nrps_pks_domains",
protein_location=prot_loc,
locus_tag=gene.get_name())
new_feature.domain = domain.hit_id
new_feature.locus_tag = gene.locus_tag or gene.get_name()
new_feature.detection = "hmmscan"
new_feature.database = "nrpspksdomains.hmm"
new_feature.evalue = domain.evalue
new_feature.score = domain.bitscore
new_feature.translation = gene.translation[domain.query_start:domain.
query_end]
domain_counts[domain.hit_id] += 1 # 1-indexed, so increment before use
domain_name = "{}_{}.{}".format(gene.get_name(), domain.hit_id,
domain_counts[domain.hit_id])
new_feature.domain_id = "nrpspksdomains_" + domain_name
new_feature.label = domain_name
new_features[domain] = new_feature
return new_features