def c_to_p(self, var_c, pro_ac=None):
"""
Converts a c. SequenceVariant to a p. SequenceVariant on the specified protein accession
Author: Rudy Rico
:param SequenceVariant var_c: hgvsc tag
:param str pro_ac: protein accession
:rtype: hgvs.sequencevariant.SequenceVariant
"""
if not (var_c.type == "c"):
raise HGVSInvalidVariantError("Expected a cDNA (c.) variant; got " + str(var_c))
if self._validator:
self._validator.validate(var_c)
reference_data = RefTranscriptData(self.hdp, var_c.ac, pro_ac)
builder = altseqbuilder.AltSeqBuilder(var_c, reference_data)
# TODO: handle case where you get 2+ alt sequences back;
# currently get list of 1 element loop structure implemented
# to handle this, but doesn't really do anything currently.
all_alt_data = builder.build_altseq()
var_ps = []
for alt_data in all_alt_data:
builder = altseq_to_hgvsp.AltSeqToHgvsp(reference_data, alt_data)
var_p = builder.build_hgvsp()
var_ps.append(var_p)
var_p = var_ps[0]
if self.add_gene_symbol:
self._update_gene_symbol(var_p, var_c.gene)
return var_p
def _run_comparison(self, hgvsc, expected_sequence):
ac_p = "DUMMY"
var = self._parser.parse_hgvs_variant(hgvsc)
transcript_data = RefTranscriptData(hdp=self._datasource, tx_ac=var.ac, pro_ac=ac_p)
builder = altseqbuilder.AltSeqBuilder(var, transcript_data)
insert_result = builder.build_altseq()
actual_sequence = insert_result[0].transcript_sequence
msg = "expected: {}\nactual : {}".format(expected_sequence, actual_sequence)
self.assertEqual(expected_sequence, actual_sequence, msg)
def _run_comparison(self, hgvsc, expected_sequence):
# test replicates the internal class of p_to_c
@attr.s(slots=True)
class RefTranscriptData(object):
transcript_sequence = attr.ib()
aa_sequence = attr.ib()
cds_start = attr.ib()
cds_stop = attr.ib()
protein_accession = attr.ib()
@classmethod
def setup_transcript_data(cls, ac, ac_p, db, ref="GRCh37.p10"):
"""helper for generating RefTranscriptData from for c_to_p"""
tx_info = db.get_tx_info(ac)
tx_seq = db.get_tx_seq(ac)
if tx_info is None or tx_seq is None:
raise hgvs.exceptions.HGVSError("Missing transcript data for accession: {}".format(ac))
# use 1-based hgvs coords
cds_start = tx_info["cds_start_i"] + 1
cds_stop = tx_info["cds_end_i"]
# padding list so biopython won't complain during the conversion
tx_seq_to_translate = tx_seq[cds_start - 1:cds_stop]
if len(tx_seq_to_translate) % 3 != 0:
"".join(list(tx_seq_to_translate).extend(["N"] * ((3 - len(tx_seq_to_translate) % 3) % 3)))
tx_seq_cds = Seq(tx_seq_to_translate)
protein_seq = str(tx_seq_cds.translate())
transcript_data = RefTranscriptData(tx_seq, protein_seq, cds_start, cds_stop, ac_p)
return transcript_data
ac_p = "DUMMY"
var = self._parser.parse_hgvs_variant(hgvsc)
transcript_data = RefTranscriptData.setup_transcript_data(var.ac, ac_p, self._datasource)
builder = altseqbuilder.AltSeqBuilder(var, transcript_data)
insert_result = builder.build_altseq()
actual_sequence = insert_result[0].transcript_sequence
msg = "expected: {}\nactual : {}".format(expected_sequence, actual_sequence)
self.assertEqual(expected_sequence, actual_sequence, msg)
def c_to_p(self, var_c, pro_ac=None):
"""
Converts a c. SequenceVariant to a p. SequenceVariant on the specified protein accession
Author: Rudy Rico
:param SequenceVariant var_c: hgvsc tag
:param str pro_ac: protein accession
:rtype: hgvs.sequencevariant.SequenceVariant
"""
@attr.s(slots=True)
class RefTranscriptData(object):
transcript_sequence = attr.ib()
aa_sequence = attr.ib()
cds_start = attr.ib()
cds_stop = attr.ib()
protein_accession = attr.ib()
@classmethod
def setup_transcript_data(cls, hdp, tx_ac, pro_ac):
"""helper for generating RefTranscriptData from for c_to_p"""
tx_info = hdp.get_tx_identity_info(var_c.ac)
tx_seq = hdp.get_seq(tx_ac)
if tx_info is None or tx_seq is None:
raise HGVSDataNotAvailableError("Missing transcript data for accession: {}".format(tx_ac))
# use 1-based hgvs coords
cds_start = tx_info["cds_start_i"] + 1
cds_stop = tx_info["cds_end_i"]
# padding list so biopython won't complain during the conversion
tx_seq_to_translate = tx_seq[cds_start - 1:cds_stop]
if len(tx_seq_to_translate) % 3 != 0:
"".join(list(tx_seq_to_translate).extend(["N"] * ((3 - len(tx_seq_to_translate) % 3) % 3)))
tx_seq_cds = Seq(tx_seq_to_translate)
protein_seq = str(tx_seq_cds.translate())
if pro_ac is None:
# get_acs... will always return at least the MD5_ accession
pro_ac = (hdp.get_pro_ac_for_tx_ac(tx_ac) or hdp.get_acs_for_protein_seq(protein_seq)[0])
transcript_data = RefTranscriptData(tx_seq, protein_seq, cds_start, cds_stop, pro_ac)
return transcript_data
if not (var_c.type == "c"):
raise HGVSInvalidVariantError("Expected a cDNA (c.); got " + str(var_c))
if self._validator:
self._validator.validate(var_c)
reference_data = RefTranscriptData.setup_transcript_data(self.hdp, var_c.ac, pro_ac)
builder = altseqbuilder.AltSeqBuilder(var_c, reference_data)
# TODO: handle case where you get 2+ alt sequences back;
# currently get list of 1 element loop structure implemented
# to handle this, but doesn't really do anything currently.
all_alt_data = builder.build_altseq()
var_ps = []
for alt_data in all_alt_data:
builder = altseq_to_hgvsp.AltSeqToHgvsp(reference_data, alt_data)
var_p = builder.build_hgvsp()
var_ps.append(var_p)
var_p = var_ps[0]
return var_p
