def parse_mhc_allele(self,
allele: str,
pattern=H2_ALLELE_PATTERN) -> MhcAllele:
match = H2_NETMHCPAN_ALLELE_PATTERN.match(allele)
if match:
# this ensures that netmhcpan output is normalized
allele = "H2{gene}{protein}".format(gene=match.group(1),
protein=match.group(2))
match = H2_ALLELE_PATTERN.match(allele)
if match is None:
raise NeofoxDataValidationException(
"Allele does not match H2 allele pattern {}".
format(allele) if allele != "" else
"Please check the format of provided alleles. An empty allele is provided"
)
gene = match.group(1)
protein = match.group(2)
# controls for existence in the HLA database and warns the user
mhc_allele = MhcAllele(gene=gene, protein=protein)
if not self.mhc_database.exists(mhc_allele):
logger.warning(
"Allele {} does not exist in the H2 database".format(allele))
# builds a normalized representation of the allele
name = "{gene}{protein}".format(gene=gene, protein=protein)
# full name is the same as name in this case as the pattern does not allow variability
mhc_allele.name = name
mhc_allele.full_name = name
return mhc_allele
def parse_mhc_allele(self, allele: str) -> MhcAllele:
match = HLA_ALLELE_PATTERN_WITHOUT_SEPARATOR.match(allele)
if match is not None:
# allele without separator, controls for ambiguities
gene = match.group(1)
group = match.group(2)
protein = match.group(3)
default_allele_exists = self.mhc_database.exists(
MhcAllele(gene=gene, group=group, protein=protein))
if not default_allele_exists:
# if default allele does not exist, tries alternative
protein = group[-1:] + protein
group = group[0:-1]
else:
# infers gene, group and protein from the name
match = HLA_ALLELE_PATTERN.match(allele)
if match is None:
raise NeofoxDataValidationException(
"Allele does not match HLA allele pattern {}".
format(allele) if allele != "" else
"Please check the format of provided alleles. An empty allele is provided"
)
gene = match.group(1)
group = match.group(2)
protein = match.group(3)
# controls for existence in the HLA database and warns the user
mhc_allele = MhcAllele(gene=gene, group=group, protein=protein)
if not self.mhc_database.exists(mhc_allele):
logger.warning(
"Allele {} does not exist in the HLA database".format(allele))
# builds a normalized representation of the allele
name = "HLA-{gene}*{serotype}:{protein}".format(gene=gene,
serotype=group,
protein=protein)
# ensures that full name stores the complete allele as provided but normalizes
# its representation
full_name = name
six_digits_id = match.group(4)
if six_digits_id is not None and six_digits_id != "":
full_name = full_name + ":{}".format(six_digits_id)
eight_digits_id = match.group(5)
if eight_digits_id is not None and eight_digits_id != "":
full_name = full_name + ":{}".format(eight_digits_id)
expression_change = match.group(6)
if expression_change is not None and expression_change != "":
full_name = full_name + expression_change
mhc_allele.name = name
mhc_allele.full_name = full_name
return mhc_allele