diff cravat_convert/cravat_convert.py @ 14:73ed52b25298 draft

Uploaded
author in_silico
date Wed, 13 Jun 2018 12:21:05 -0400
parents
children
line wrap: on
line diff
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/cravat_convert/cravat_convert.py	Wed Jun 13 12:21:05 2018 -0400
@@ -0,0 +1,86 @@
+'''
+Convert a VCF format file to Cravat format file
+'''
+
+import os
+import argparse
+from vcf_converter import CravatConverter
+###
+# import ipdb
+
+
+def get_vcf_mapping():
+    """ : VCF Headers mapped to their index position in a row of VCF values.
+        : These are only the mandatory columns, per the VCF spec.
+    """
+    return {
+        'CHROM': 0,
+        'POS': 1,
+        'ID': 2,
+        'REF': 3,
+        'ALT': 4,
+        'QUAL': 5,
+        'FILTER': 6,
+        'INFO': 7
+    }
+
+
+def get_args():
+    parser = argparse.ArgumentParser()
+    parser.add_argument('--input',
+                            '-i',
+                            required = True,
+                            help='Input path to a VCF file for conversion',)
+    parser.add_argument('--output',
+                            '-o',
+                            default = None,
+                            help = 'Output path to write the cravat file to')
+    return parser.parse_args()
+
+
+def convert(in_path, out_path=None, cr_sep='\t', cr_newline='\n'):
+    """ : Convert a VCF file to a Cravat file.
+        : Arguments:
+            : in_path: <str> path to input vcf file
+            : out_path: <str> path to output cravat file. Will defualt to cravat_converted.txt in the input directory.
+            : cr_sep: <str> the value delimiter for the output cravat file. Default value of '\\t'.
+            : out_newline: <str> the newline delimiter in the output cravat file. Default of '\\n'
+    """
+    if not out_path:
+        base, _ = os.path.split(in_path)
+        out_path = os.path.join(base, "cravat_converted.txt")
+    
+    with open(in_path, 'r') as in_file, \
+    open(out_path, 'w') as out_file:
+
+        # cr_count will be used to generate the 'TR' field of the cravat rows (first header)
+        cr_count = 0
+        # VCF lines are always assumed to be '+' strand, as VCF doesn't specify that attribute
+        strand = '+'
+        # VCF converter. Adjusts position, reference, and alternate for Cravat formatting.
+        converter = CravatConverter()
+        # A dictionary of mandatory vcf headers mapped to their row indices
+        vcf_mapping = get_vcf_mapping()
+
+        for line in in_file:
+            if line.startswith("#"):
+                continue
+            line = line.strip().split()
+            # row is dict of VCF headers mapped to corresponding values of this line
+            row = { header: line[index] for header, index in vcf_mapping.items() }
+            for alt in row["ALT"].split(","):
+                new_pos, new_ref, new_alt = converter.extract_vcf_variant(strand, row["POS"], row["REF"], alt)
+                new_pos, new_ref, new_alt = str(new_pos), str(new_ref), str(new_alt)
+                cr_line = cr_sep.join([
+                    'TR' + str(cr_count), row['CHROM'], new_pos, strand, new_ref, new_alt, row['ID']
+                ])
+                out_file.write(cr_line + cr_newline)
+                cr_count += 1
+
+
+if __name__ == "__main__":
+    cli_args = get_args()
+    if cli_args.output == None:
+        base, _ = os.path.split(cli_args.input)
+        cli_args.output = os.path.join(base, "cravat_converted.txt") 
+    convert(in_path = cli_args.input, out_path = cli_args.output)