diff cravat_annotate/cravat_annotate.py @ 22:e05043fbb945 draft

Uploaded
author in_silico
date Wed, 27 Jun 2018 17:53:25 -0400
parents
children
line wrap: on
line diff
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/cravat_annotate/cravat_annotate.py	Wed Jun 27 17:53:25 2018 -0400
@@ -0,0 +1,128 @@
+"""
+A galaxy wrapper for the /rest/service/query API endpoint on Cravat.
+"""
+
+
+import requests
+import json
+import sys
+import re
+import argparse
+import ipdb
+
+
+# The endpoint that CravatQuerys are submitted to
+endpoint = 'http://www.cravat.us/CRAVAT/rest/service/query'
+
+
+# The value delimiter used in the Cravat input file to delimit values
+delimiter = "\t"
+
+
+# Defualt indices for intepretting a cravat file's row of data in to a CravatQuery
+cr_mapping = {
+	'chromosome': 1,
+	'position': 2,
+	'strand': 3,
+	'reference': 4,
+	'alternate': 5
+}
+
+
+# The neccessary attributes neeeded to submit a query.
+query_keys = [
+	'chromosome', 'position', 'strand', 'reference', 'alternate'
+]
+
+
+# Expected response keys from server. Ordered in list so that galaxy output has uniform column ordering run-to-run.
+# If cravat server returns additional keys, they are appended to and included in output.
+response_keys = [
+	"Chromosome", "Position", "Strand", "Reference base(s)", "Alternate base(s)",
+	"HUGO symbol", "S.O. transcript", "Sequence ontology protein change", "Sequence ontology",
+	"S.O. all transcripts", "gnomAD AF", "gnomAD AF (African)", "gnomAD AF (Amrican)",
+	"gnomAD AF (Ashkenazi Jewish)", "gnomAD AF (East Asian)", "gnomAD AF (Finnish)",
+	"gnomAD AF (Non-Finnish European)", "gnomAD AF (Other)", "gnomAD AF (South Asian)",
+	"1000 Genomes AF", "ESP6500 AF (average)", "ESP6500 AF (European American)",
+	"ESP6500 AF (African American)", "COSMIC transcript", "COSMIC protein change", 
+	"COSMIC variant count [exact nucleotide change]", "cosmic_site_nt", "CGL driver class",
+	"TARGET", "dbSNP", "cgc_role", "cgc_inheritance", "cgc_tumor_type_somatic",
+	"cgc_tumor_type_germline", "ClinVar", "ClinVar disease identifier", "ClinVar XRef",
+	"GWAS Phenotype (GRASP)", "GWAS PMID (GRASP)", "Protein 3D variant"
+]
+
+
+def get_args():
+    parser = argparse.ArgumentParser()
+    parser.add_argument('--input',
+                            '-i',
+                            required = True,
+                            help='Input path to a cravat file for querying',)
+    parser.add_argument('--output',
+                            '-o',
+                            default = None,
+                            help = 'Output path to write results from query')
+    return parser.parse_args()
+
+
+def format_chromosome(chrom):
+	""" : Ensure chromosome entry is propely formatted for use as querying attribute. """
+	if chrom[0:3] == 'chr':
+		return chrom
+	return 'chr' + str(chrom)
+
+
+def get_query_string(row):
+	""" : From a row dict, return a query string for the Cravat server.
+		: The row dict is cravat headeres associated to their values of that row.
+	"""
+	return '_'.join([ row['chromosome'], row['position'], row['strand'], row['reference'], row['alternate'] ])
+
+
+def query(in_path, out_path):
+	""" : From a Cravat the file at in_path, query each line on the Cravat server.
+		: Write the response values to file at out_path.
+	"""
+
+	with open(in_path, 'r') as in_file, \
+	open(out_path, 'w') as out_file:
+
+		for line in in_file:
+			line = line.strip().split('\t')
+			# row is dict of cravat col headers assioted values in this line
+			row = { header: line[index] for header, index in cr_mapping.items() }
+			row['chromosome'] = format_chromosome(row['chromosome'])
+			query_string = get_query_string(row)
+			call = requests.get(endpoint, params={ 'mutation': query_string })
+			if call.status_code != 200 or call.text == "":
+				raise requests.RequestException("Bad Server Response. Respone code: '{}', Response Text: '{}'".format(call.status_code, call.text))
+			json_response = json.loads(call.text)
+			# See if server returned additional json key-val paris not expected in response_keys
+			for key in json_response:
+				if key not in response_keys:
+					response_keys.append(key)
+			# Write key in order of response_keys to standardize order of output columns
+			wrote = False
+			for key in response_keys:
+				if key not in json_response:
+					val = None
+				val = json_response[key]
+				# Format standardization for numerics
+				try:
+					val = float(val)
+					val = format(val, ".4f")
+				except:
+					pass
+				if wrote:
+					out_file.write("\t")
+				out_file.write(val)
+				wrote = True
+			out_file.write("\n")		
+
+
+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") 
+	query(cli_args.input, cli_args.output)