def delete(ctx, variant_file, family_file, family_type, family_id): """Delete the variants of a case.""" if not family_file or family_id: logger.error("Please provide a family file or a case id") logger.info("Exiting") ctx.abort() family = get_family( family_lines=family_file, family_type=family_type ) family_id = family.family_id affected_individuals = family.affected_individuals adapter = ctx.obj['adapter'] if variant_file == '-': logger.info("Start parsing variants from stdin") variant_stream = get_vcf_handle(fsock=sys.stdin) else: logger.info("Start parsing variants from stdin") variant_stream = get_vcf_handle(infile=variant_file) start_deleting = datetime.now() try: count = delete_variants(adapter, variant_stream, family_id, affected_individuals) except CaseError as error: logger.warning(error.message) ctx.abort() logger.info("Nr of variants deleted: {0}".format(count)) logger.info("Time to delete variants: {0}" .format(datetime.now() - start_deleting))
def get_header(vcf_file_path): """Parse the header and return a header object Args: vcf_file_path(str): Path to vcf Returns: head: A HeaderParser object """ logger.info("Parsing header of file {0}".format(vcf_file_path)) head = HeaderParser() handle = get_vcf_handle(infile=vcf_file_path) # Parse the header for line in handle: line = line.rstrip() if line.startswith('#'): if line.startswith('##'): head.parse_meta_data(line) else: head.parse_header_line(line) else: break handle.close() return head
def variant(self, case_id, variant_id): """Return a specific variant. Args: case_id (str): Path to vcf file variant_id (str): A variant id Returns: variant (Variant): The variant object for the given id """ case_obj = self.case(case_id=case_id) vcf_file_path = case_obj.variant_source head = self._get_header(vcf_file_path) handle = get_vcf_handle(infile=vcf_file_path) relevant_lines = (line for line in handle if not line.startswith('#')) for index, variant_line in enumerate(relevant_lines): index += 1 line_id = get_variant_id(variant_line=variant_line).lstrip('chrCHR') if line_id == variant_id: return self._format_variant( variant_line=variant_line, index=index, case_obj=case_obj, head=head ) return None
def test_get_vcf_handle_file(self): """docstring for test_get_vcf_handle_file""" file_handle = get_vcf_handle(infile=self.temp_file.name) result = [] for line in file_handle: line = line.rstrip() result.append(line) assert result == vcf_lines
def _get_filtered_variants(self, vcf_file_path, filters={}): """Check if variants follows the filters This function will try to make filters faster for the vcf adapter Args: vcf_file_path(str): Path to vcf filters (dict): A dictionary with filters Yields: varian_line (str): A vcf variant line """ genes = set() consequences = set() sv_types = set() if filters.get('gene_ids'): genes = set([gene_id.strip() for gene_id in filters['gene_ids']]) if filters.get('consequence'): consequences = set(filters['consequence']) if filters.get('sv_types'): sv_types = set(filters['sv_types']) logger.info("Get variants from {0}".format(vcf_file_path)) handle = get_vcf_handle(infile=vcf_file_path) for variant_line in handle: if not variant_line.startswith('#'): keep_variant = True if genes and keep_variant: keep_variant = False for gene in genes: if "{0}".format(gene) in variant_line: keep_variant = True break if consequences and keep_variant: keep_variant = False for consequence in consequences: if consequence in variant_line: keep_variant = True break if sv_types and keep_variant: keep_variant = False for sv_type in sv_types: if sv_type in variant_line: keep_variant = True break if keep_variant: yield variant_line
def _get_filtered_variants(self, case_obj, filters={}): """Check if variants follows the filters This function will try to make filters faster for the vcf adapter Args: case_obj (puzzle.models.Case): A case object filters (dict): A dictionary with filters """ genes = set() consequences = set() sv_types = set() vcf_file_path = case_obj.variant_source logger.info("Parsing file {0}".format(vcf_file_path)) if filters.get("gene_ids"): genes = set([gene_id.strip() for gene_id in filters["gene_ids"]]) if filters.get("consequence"): consequences = set(filters["consequence"]) if filters.get("sv_types"): sv_types = set(filters["sv_types"]) handle = get_vcf_handle(infile=vcf_file_path) for variant_line in handle: if not variant_line.startswith("#"): keep_variant = True if genes and keep_variant: keep_variant = False for gene in genes: if "|{0}|".format(gene) in variant_line: keep_variant = True break if consequences and keep_variant: keep_variant = False for consequence in consequences: if consequence in variant_line: keep_variant = True break if sv_types and keep_variant: keep_variant = False for sv_type in sv_types: if sv_type in variant_line: keep_variant = True break if keep_variant: yield variant_line
def load(ctx, variant_file, family_file, family_type, bulk_insert): """Load the variants of a case The loading is based on if the variant is seen in a ny affected individual in the family. """ if not family_file: logger.error("Please provide a family file") ctx.abort() if variant_file == '-': logger.info("Parsing variants from stdin") variant_file = get_vcf_handle(fsock=sys.stdin) else: logger.info("Start parsing variants from stdin") variant_path = os.path.abspath(variant_file) variant_file = get_vcf_handle(infile=variant_file) try: family = get_family(family_lines=family_file, family_type=family_type) except SyntaxError as error: logger.warning(error.message) ctx.abort() if not family.affected_individuals: logger.error("No affected individuals could be found in ped file") ctx.abort() logger.info("Found affected individuals in ped file: {0}" .format(', '.join(family.affected_individuals))) adapter = ctx.obj['adapter'] try: load_variants(adapter, family.family_id, family.affected_individuals, variant_file, bulk_insert=bulk_insert, vcf_path=variant_path) except CaseError as error: logger.error(error.message) ctx.abort()
def _formated_variants(self, raw_variants, case_obj): """Return variant objects Args: raw_variants (Iterable): An iterable with variant lines case_obj (puzzle.nodels.Case): A case object """ vcf_file_path = case_obj.variant_source logger.info("Parsing file {0}".format(vcf_file_path)) head = HeaderParser() handle = get_vcf_handle(infile=vcf_file_path) # Parse the header for line in handle: line = line.rstrip() if line.startswith("#"): if line.startswith("##"): head.parse_meta_data(line) else: head.parse_header_line(line) else: break handle.close() header_line = head.header # Get the individual ids for individuals in vcf file vcf_individuals = set([ind_id for ind_id in head.individuals]) variant_columns = ["CHROM", "POS", "ID", "REF", "ALT", "QUAL", "FILTER"] vep_header = head.vep_columns snpeff_header = head.snpeff_columns index = 0 for variant_line in raw_variants: if not variant_line.startswith("#"): index += 1 # Create a variant dict: variant_dict = get_variant_dict(variant_line=variant_line, header_line=header_line) variant_dict["CHROM"] = variant_dict["CHROM"].lstrip("chrCHR") # Crreate a info dict: info_dict = get_info_dict(info_line=variant_dict["INFO"]) # Check if vep annotation: vep_string = info_dict.get("CSQ") # Check if snpeff annotation: snpeff_string = info_dict.get("ANN") if vep_string: # Get the vep annotations vep_info = get_vep_info(vep_string=vep_string, vep_header=vep_header) elif snpeff_string: # Get the vep annotations snpeff_info = get_snpeff_info(snpeff_string=snpeff_string, snpeff_header=snpeff_header) variant = Variant(**{column: variant_dict.get(column, ".") for column in variant_columns}) logger.debug("Creating a variant object of variant {0}".format(variant.get("variant_id"))) variant["index"] = index logger.debug("Updating index to: {0}".format(index)) 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"] and not "<" 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") else: variant["stop"] = int(info_dict.get("END", variant_dict["POS"])) variant["sv_len"] = variant["stop"] - variant["start"] variant["stop_chrom"] = other_chrom else: variant["stop"] = int(variant_dict["POS"]) + (len(variant_dict["REF"]) - len(variant_dict["ALT"])) variant["sv_type"] = info_dict.get("SVTYPE") 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"]) # It would be easy to update these keys... thousand_g = info_dict.get("1000GAF") if thousand_g: logger.debug("Updating thousand_g to: {0}".format(thousand_g)) variant["thousand_g"] = float(thousand_g) variant.add_frequency("1000GAF", variant.get("thousand_g")) # SV specific tag for number of occurances occurances = info_dict.get("OCC") if occurances: logger.debug("Updating occurances to: {0}".format(occurances)) variant["occurances"] = float(occurances) variant.add_frequency("OCC", occurances) cadd_score = info_dict.get("CADD") if cadd_score: logger.debug("Updating cadd_score to: {0}".format(cadd_score)) variant["cadd_score"] = float(cadd_score) rank_score_entry = info_dict.get("RankScore") if rank_score_entry: for family_annotation in rank_score_entry.split(","): rank_score = family_annotation.split(":")[-1] logger.debug("Updating rank_score to: {0}".format(rank_score)) variant["rank_score"] = float(rank_score) genetic_models_entry = info_dict.get("GeneticModels") if genetic_models_entry: genetic_models = [] for family_annotation in genetic_models_entry.split(","): for genetic_model in family_annotation.split(":")[-1].split("|"): genetic_models.append(genetic_model) logger.debug("Updating rank_score to: {0}".format(rank_score)) variant["genetic_models"] = genetic_models # Add genotype calls: for individual in case_obj.individuals: sample_id = individual.ind_id if sample_id in vcf_individuals: raw_call = dict(zip(variant_dict["FORMAT"].split(":"), variant_dict[sample_id].split(":"))) variant.add_individual( Genotype( sample_id=sample_id, genotype=raw_call.get("GT", "./."), case_id=individual.case_name, phenotype=individual.phenotype, ref_depth=raw_call.get("AD", ",").split(",")[0], alt_depth=raw_call.get("AD", ",").split(",")[1], genotype_quality=raw_call.get("GQ", "."), depth=raw_call.get("DP", "."), supporting_evidence=raw_call.get("SU", "0"), pe_support=raw_call.get("PE", "0"), sr_support=raw_call.get("SR", "0"), ) ) # Add transcript information: gmaf = None if vep_string: for transcript_info in vep_info: transcript = self._get_vep_transcripts(transcript_info) gmaf_raw = transcript_info.get("GMAF") if gmaf_raw: gmaf = float(gmaf_raw.split(":")[-1]) variant.add_transcript(transcript) if gmaf: variant.add_frequency("GMAF", gmaf) if not variant.thousand_g: variant.thousand_g = gmaf elif snpeff_string: for transcript_info in snpeff_info: transcript = self._get_snpeff_transcripts(transcript_info) variant.add_transcript(transcript) variant["most_severe_consequence"] = get_most_severe_consequence(variant["transcripts"]) for gene in self._get_genes(variant): variant.add_gene(gene) self._add_compounds(variant=variant, info_dict=info_dict) yield variant
def get_individuals(vcf=None, case_lines=None, case_type='ped'): """Get the individuals from a vcf file, and/or a ped file. Args: vcf (str): Path to a vcf case_lines(Iterable): Ped like lines case_type(str): Format of ped lines Returns: individuals (generator): generator with Individuals """ individuals = [] if case_lines: # read individuals from ped file family_parser = FamilyParser(case_lines, family_type=case_type) families = family_parser.families logger.info("Found families {0}".format( ','.join(list(families.keys())))) if len(families) != 1: logger.error("Only one family can be used with vcf adapter") raise IOError case_id = list(families.keys())[0] logger.info("Family used in analysis: {0}".format(case_id)) for ind_id in family_parser.individuals: ind = family_parser.individuals[ind_id] logger.info("Found individual {0}".format(ind.individual_id)) individual = Individual( ind_id=ind.individual_id, case_id=case_id, mother=ind.mother, father=ind.father, sex=str(ind.sex), phenotype=str(ind.phenotype), variant_source=vcf, ) individuals.append(individual) elif vcf: # read individuals from vcf file case_id = os.path.basename(vcf) head = HeaderParser() handle = get_vcf_handle(infile=vcf) for line in handle: line = line.rstrip() if line.startswith('#'): if line.startswith('##'): head.parse_meta_data(line) else: head.parse_header_line(line) else: break for index, ind in enumerate(head.individuals): # If we only have a vcf file we can not get metadata about the # individuals individual = Individual( ind_id=ind, case_id=case_id, variant_source=vcf, ) individuals.append(individual) logger.debug("Found individual {0} in {1}".format( ind, vcf)) return individuals
def test_get_vcf_handle_no_input(self): """docstring for test_get_vcf_handle_file""" with pytest.raises(IOError): file_handle = get_vcf_handle()
def _formated_variants(self, raw_variants, case_obj): """Return variant objects Args: raw_variants (Iterable): An iterable with variant lines case_obj (puzzle.nodels.Case): A case object """ vcf_file_path = case_obj.variant_source logger.info("Parsing file {0}".format(vcf_file_path)) head = HeaderParser() handle = get_vcf_handle(infile=vcf_file_path) # Parse the header for line in handle: line = line.rstrip() if line.startswith('#'): if line.startswith('##'): head.parse_meta_data(line) else: head.parse_header_line(line) else: break handle.close() header_line = head.header # Get the individual ids for individuals in vcf file vcf_individuals = set([ind_id for ind_id in head.individuals]) variant_columns = ['CHROM', 'POS', 'ID', 'REF', 'ALT', 'QUAL', 'FILTER'] vep_header = head.vep_columns snpeff_header = head.snpeff_columns index = 0 for variant_line in raw_variants: if not variant_line.startswith('#'): index += 1 #Create a variant dict: variant_dict = get_variant_dict( variant_line = variant_line, header_line = header_line ) variant_dict['CHROM'] = variant_dict['CHROM'].lstrip('chrCHR') #Crreate a info dict: info_dict = get_info_dict( info_line = variant_dict['INFO'] ) #Check if vep annotation: vep_string = info_dict.get('CSQ') #Check if snpeff annotation: snpeff_string = info_dict.get('ANN') if vep_string: #Get the vep annotations vep_info = get_vep_info( vep_string = vep_string, vep_header = vep_header ) elif snpeff_string: #Get the vep annotations snpeff_info = get_snpeff_info( snpeff_string = snpeff_string, snpeff_header = snpeff_header ) variant = Variant( **{column: variant_dict.get(column, '.') for column in variant_columns} ) logger.debug("Creating a variant object of variant {0}".format( variant.get('variant_id'))) variant['index'] = index logger.debug("Updating index to: {0}".format( index)) 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') else: variant['stop'] = int(info_dict.get('END', variant_dict['POS'])) variant['sv_len'] = variant['stop'] - variant['start'] variant['stop_chrom'] = other_chrom else: variant['stop'] = int(variant_dict['POS']) + \ (len(variant_dict['REF']) - len(variant_dict['ALT'])) variant['sv_type'] = info_dict.get('SVTYPE') 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']) # It would be easy to update these keys... thousand_g = info_dict.get('1000GAF') if thousand_g: logger.debug("Updating thousand_g to: {0}".format( thousand_g)) variant['thousand_g'] = float(thousand_g) variant.add_frequency('1000GAF', variant.get('thousand_g')) #SV specific tag for number of occurances occurances = info_dict.get('OCC') if occurances: logger.debug("Updating occurances to: {0}".format( occurances)) variant['occurances'] = float(occurances) variant.add_frequency('OCC', occurances) cadd_score = info_dict.get('CADD') if cadd_score: logger.debug("Updating cadd_score to: {0}".format( cadd_score)) variant['cadd_score'] = float(cadd_score) rank_score_entry = info_dict.get('RankScore') if rank_score_entry: for family_annotation in rank_score_entry.split(','): rank_score = family_annotation.split(':')[-1] logger.debug("Updating rank_score to: {0}".format( rank_score)) variant['rank_score'] = float(rank_score) genetic_models_entry = info_dict.get('GeneticModels') if genetic_models_entry: genetic_models = [] for family_annotation in genetic_models_entry.split(','): for genetic_model in family_annotation.split(':')[-1].split('|'): genetic_models.append(genetic_model) logger.debug("Updating rank_score to: {0}".format( rank_score)) variant['genetic_models'] = genetic_models #Add genotype calls: for individual in case_obj.individuals: sample_id = individual.ind_id if sample_id in vcf_individuals: raw_call = dict(zip( variant_dict['FORMAT'].split(':'), variant_dict[sample_id].split(':')) ) variant.add_individual(Genotype( sample_id = sample_id, genotype = raw_call.get('GT', './.'), case_id = individual.case_name, phenotype = individual.phenotype, ref_depth = raw_call.get('AD', ',').split(',')[0], alt_depth = raw_call.get('AD', ',').split(',')[1], genotype_quality = raw_call.get('GQ', '.'), depth = raw_call.get('DP', '.'), supporting_evidence = raw_call.get('SU', '0'), pe_support = raw_call.get('PE', '0'), sr_support = raw_call.get('SR', '0'), )) # Add transcript information: if vep_string: for transcript in self._get_vep_transcripts(variant, vep_info): variant.add_transcript(transcript) elif snpeff_string: for transcript in self._get_snpeff_transcripts(variant, snpeff_info): variant.add_transcript(transcript) variant['most_severe_consequence'] = get_most_severe_consequence( variant['transcripts'] ) for gene in self._get_genes(variant): variant.add_gene(gene) self._add_compounds(variant=variant, info_dict=info_dict) yield variant