Added alternative to choose allele symbol when splitting variants

scripts/vcf_parser

@@ -59,7 +59,11 @@ def print_version(ctx, param, value):
                     type=click.Path(exists=False),
                     help='Path to a outfile.'
 )
-@click.option('-v', '--verbose', 
+@click.option("--allele_symbol",
+                default='0',
+                help="The symbol that should be used when representing "\
+                "unobserved alleles. Default is '0'")
+@click.option('-v', '--verbose',
                 is_flag=True,
                 help='Increase output verbosity.'
 )
@@ -88,7 +92,7 @@ def print_version(ctx, param, value):
                     help="Set the level of log output."
 )
 def cli(variant_file, vep, split, outfile, verbose, silent, skip_info_check,
-        logfile, loglevel):
+        allele_symbol, logfile, loglevel):
     """
     Tool for parsing vcf files.
     
@@ -107,11 +111,11 @@ def cli(variant_file, vep, split, outfile, verbose, silent, skip_info_check,
     if variant_file == '-':
         logger.info("Start parsing variants from stdin")
         my_parser = vcf_parser.VCFParser(fsock=sys.stdin, split_variants=split, 
-                                            skip_info_check=skip_info_check)
+                    skip_info_check=skip_info_check, allele_symbol=allele_symbol)
     else:
         logger.info("Start parsing variants from file {0}".format(variant_file))
         my_parser = vcf_parser.VCFParser(infile = variant_file, 
-        split_variants=split, skip_info_check=skip_info_check)
+        split_variants=split, skip_info_check=skip_info_check, allele_symbol=allele_symbol)
 
     start = datetime.now()
     nr_of_variants = 0

vcf_parser/parser.py

@@ -86,7 +86,8 @@
 
 class VCFParser(object):
     """docstring for VCFParser"""
-    def __init__(self, infile=None, fsock=None, split_variants=False, skip_info_check=False):
+    def __init__(self, infile=None, fsock=None, split_variants=False, 
+                skip_info_check=False, allele_symbol='0'):
         super(VCFParser, self).__init__()
         self.logger = logging.getLogger(__name__)
 
@@ -99,6 +100,9 @@ def __init__(self, infile=None, fsock=None, split_variants=False, skip_info_chec
 
         self.skip_info_check = skip_info_check
         self.logger.info("Skip info check = {0}".format(self.skip_info_check))
+
+        self.allele_symbol = allele_symbol
+        self.logger.info("Allele symbol = {0}".format(self.allele_symbol))
 
         self.logger.info("Initializing HeaderParser")
         self.metadata = HeaderParser()
@@ -166,7 +170,8 @@ def add_variant(self, chrom, pos, rs_id, ref, alt, qual, filt, info, form=None,
 
         # If multiple alternative and split_variants we must split the variant                 
         else:
-            for splitted_variant in split_variants(variant, self.metadata):
+            for splitted_variant in split_variants(variant_dict=variant, 
+            header_parser=self.metadata, allele_symbol=self.allele_symbol):
                 self.variants.append(splitted_variant)
 
 
@@ -185,7 +190,8 @@ def __iter__(self):
                 if not (self.split_variants and len(first_variant['ALT'].split(',')) > 1):
                     variants.append(first_variant)
                 else:
-                    for splitted_variant in split_variants(first_variant, self.metadata):
+                    for splitted_variant in split_variants(variant_dict=first_variant, 
+                    header_parser=self.metadata, allele_symbol=self.allele_symbol):
                         variants.append(splitted_variant)
 
                 for variant in variants:
@@ -208,7 +214,10 @@ def __iter__(self):
                         variants.append(variant)
 
                     else:
-                        for splitted_variant in split_variants(variant, self.metadata):
+                        for splitted_variant in split_variants(
+                                    variant_dict=variant, 
+                                    header_parser=self.metadata, 
+                                    allele_symbol=self.allele_symbol):
                             variants.append(splitted_variant)
 
                 for variant in variants:

vcf_parser/utils/split_genotype.py

@@ -17,6 +17,8 @@ def split_genotype(genotype, gt_format, alternative_number, allele_symbol = '0')
         new_genotype (str): A string that represents the new genotype
     """
     logger = getLogger(__name__)
+    logger.info("Allele symbol {0}".format(allele_symbol))
+
     splitted_genotype = genotype.split(':')
     logger.debug("Parsing genotype {0}".format(splitted_genotype))
     splitted_gt_format = gt_format.split(':')

vcf_parser/utils/split_variants.py

@@ -12,7 +12,7 @@
 from vcf_parser import Genotype
 from vcf_parser.utils import (build_vep_string, split_genotype, build_info_string)
 
-def split_variants(variant_dict, header_parser):
+def split_variants(variant_dict, header_parser, allele_symbol='0'):
     """
     Checks if there are multiple alternative alleles and splitts the 
     variant.
@@ -26,7 +26,8 @@ def split_variants(variant_dict, header_parser):
         variant: A variant dictionary with the splitted information for each
                 alternative
     """
-
+    logger = getLogger(__name__)
+    logger.info("Allele symbol {0}".format(allele_symbol))
     alternatives = variant_dict['ALT'].split(',')
     reference = variant_dict['REF']
     number_of_values = 1
@@ -102,7 +103,8 @@ def split_variants(variant_dict, header_parser):
             new_genotype = split_genotype(
                             variant_dict[individual], 
                             variant['FORMAT'], 
-                            alternative_number
+                            alternative_number, 
+                            allele_symbol
                         )
 
             variant[individual] = new_genotype