def _get_puzzle_variant(self, gemini_variant, index):
"""Take a gemini variant and return a basic puzzle variant
For the overview we only need limited variant information
"""
variant_dict = {
'CHROM':gemini_variant['chrom'].lstrip('chrCHR'),
'POS':str(gemini_variant['start']),
'ID':gemini_variant['rs_ids'],
'REF':gemini_variant['ref'],
'ALT':gemini_variant['alt'],
'QUAL':gemini_variant['qual'],
'FILTER':gemini_variant['filter']
}
variant = Variant(**variant_dict)
variant['index'] = index
# Use the gemini id for fast search
variant.update_variant_id(gemini_variant['variant_id'])
#Add the most severe consequence
variant['most_severe_consequence'] = gemini_variant['impact_so']
#Add the impact severity
variant['impact_severity'] = gemini_variant['impact_severity']
max_freq = gemini_variant['max_aaf_all']
if max_freq:
variant.set_max_freq(max_freq)
#### Check the impact annotations ####
if gemini_variant['cadd_scaled']:
variant['cadd_score'] = gemini_variant['cadd_scaled']
return variant
def _format_variant(self, gemini_variant, individual_objs, index=0):
"""Make a puzzle variant from a gemini variant
Args:
gemini_variant (GeminiQueryRow): The gemini variant
individual_objs (list(dict)): A list of Individuals
index(int): The index of the variant
Returns:
variant (dict): A Variant object
"""
variant_dict = {
'CHROM':gemini_variant['chrom'].lstrip('chrCHR'),
'POS':str(gemini_variant['start']),
'ID':gemini_variant['rs_ids'],
'REF':gemini_variant['ref'],
'ALT':gemini_variant['alt'],
'QUAL':gemini_variant['qual'],
'FILTER':gemini_variant['filter']
}
variant = Variant(**variant_dict)
variant['index'] = index
# Use the gemini id for fast search
variant.update_variant_id(gemini_variant['variant_id'])
# Update the individuals
individual_genotypes = self._get_genotypes(
gemini_variant=gemini_variant,
individual_objs=individual_objs
)
for individual in individual_genotypes:
# Add the genotype calls to the variant
variant.add_individual(individual)
for transcript in self._get_transcripts(gemini_variant):
variant.add_transcript(transcript)
#Add the most severe consequence
variant['most_severe_consequence'] = gemini_variant['impact_so']
for gene in self._get_genes(variant):
variant.add_gene(gene)
variant['start'] = int(variant_dict['POS'])
if self.variant_type == 'sv':
other_chrom = variant['CHROM']
# If we have a translocation:
if ':' in variant_dict['ALT']:
other_coordinates = variant_dict['ALT'].strip('ACGTN[]').split(':')
other_chrom = other_coordinates[0].lstrip('chrCHR')
other_position = other_coordinates[1]
variant['stop'] = other_position
#Set 'infinity' to length if translocation
variant['sv_len'] = float('inf')
variant['sv_type'] = 'BND'
else:
variant['stop'] = int(gemini_variant['end'])
variant['sv_len'] = variant['stop'] - variant['start']
variant['sv_type'] = gemini_variant['sub_type']
variant['stop_chrom'] = other_chrom
else:
variant['stop'] = int(variant_dict['POS']) + \
(len(variant_dict['REF']) - len(variant_dict['ALT']))
variant['cytoband_start'] = get_cytoband_coord(
chrom=variant['CHROM'],
pos=variant['start'])
if variant.get('stop_chrom'):
variant['cytoband_stop'] = get_cytoband_coord(
chrom=variant['stop_chrom'],
pos=variant['stop'])
#### Check the impact annotations ####
if gemini_variant['cadd_scaled']:
variant['cadd_score'] = gemini_variant['cadd_scaled']
# We use the prediction in text
polyphen = gemini_variant['polyphen_pred']
if polyphen:
variant.add_severity('Polyphen', polyphen)
# We use the prediction in text
sift = gemini_variant['sift_pred']
if sift:
variant.add_severity('SIFT', sift)
#### Check the frequencies ####
thousand_g = gemini_variant['aaf_1kg_all']
if thousand_g:
variant['thousand_g'] = float(thousand_g)
variant.add_frequency(name='1000GAF', value=float(thousand_g))
exac = gemini_variant['aaf_exac_all']
if exac:
variant.add_frequency(name='EXaC', value=float(exac))
esp = gemini_variant['aaf_esp_all']
if esp:
variant.add_frequency(name='ESP', value=float(esp))
max_freq = gemini_variant['max_aaf_all']
if max_freq:
variant.set_max_freq(max_freq)
return variant
def _format_variants(self, variant, index, case_obj, add_all_info=False):
"""Return a Variant object
Format variant make a variant that includes enough information for
the variant view.
If add_all_info then all transcripts will be parsed
Args:
variant (cython2.Variant): A variant object
index (int): The index of the variant
case_obj (puzzle.models.Case): A case object
"""
header_line = self.head.header
# Get the individual ids for individuals in vcf file
vcf_individuals = set([ind_id for ind_id in self.head.individuals])
#Create a info dict:
info_dict = dict(variant.INFO)
chrom = variant.CHROM
if chrom.startswith('chr') or chrom.startswith('CHR'):
chrom = chrom[3:]
variant_obj = Variant(
CHROM=chrom,
POS=variant.POS,
ID=variant.ID,
REF=variant.REF,
ALT=variant.ALT[0],
QUAL=variant.QUAL,
FILTER=variant.FILTER,
)
variant_obj._set_variant_id()
logger.debug("Creating a variant object of variant {0}".format(
variant_obj.variant_id))
variant_obj.index = index
logger.debug("Updating index to: {0}".format(index))
########### Get the coordinates for the variant ##############
variant_obj.start = variant.start
variant_obj.stop = variant.end
#SV variants needs to be handeled a bit different since the can be huge
#it would take to much power to parse all vep/snpeff entrys for these.
if self.variant_type == 'sv':
variant_obj.stop = int(info_dict.get('END', variant_obj.POS))
self._add_sv_coordinates(variant_obj)
variant_obj.sv_type = info_dict.get('SVTYPE')
# Special for FindSV software:
# SV specific tag for number of occurances
occurances = info_dict.get('OCC')
if occurances:
logger.debug("Updating occurances to: {0}".format(occurances))
variant_obj['occurances'] = float(occurances)
variant_obj.add_frequency('OCC', occurances)
else:
self._add_thousand_g(variant_obj, info_dict)
self._add_cadd_score(variant_obj, info_dict)
self._add_genetic_models(variant_obj, info_dict)
self._add_transcripts(variant_obj, info_dict)
self._add_exac(variant_obj, info_dict)
self._add_hgnc_symbols(variant_obj)
if add_all_info:
self._add_genotype_calls(variant_obj, str(variant), case_obj)
self._add_compounds(variant_obj, info_dict)
self._add_gmaf(variant_obj, info_dict)
self._add_genes(variant_obj)
##### Add consequences ####
self._add_consequences(variant_obj, str(variant))
self._add_most_severe_consequence(variant_obj)
self._add_impact_severity(variant_obj)
self._add_rank_score(variant_obj, info_dict)
variant_obj.set_max_freq()
return variant_obj
