Mercurial > repos > iuc > data_manager_primer_scheme_bedfiles
diff data_manager/install_primer_scheme_bedfiles.py @ 0:ae730ecdc7e7 draft
"planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/data_managers/data_manager_primer_scheme_bedfiles commit f78536f45aa5ce801251a9725e468a47de1be44a"
author | iuc |
---|---|
date | Sun, 19 Apr 2020 09:22:19 -0400 |
parents | |
children | cce5d9327cd6 |
line wrap: on
line diff
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/data_manager/install_primer_scheme_bedfiles.py Sun Apr 19 09:22:19 2020 -0400 @@ -0,0 +1,141 @@ +#!/usr/bin/env python + +from __future__ import print_function, division + +import argparse +import json +import os +import os.path +import re +import sys +import shutil + +import requests + +DATA_TABLE_NAME = "primer_scheme_bedfiles" + + +def fetch_artic_primers(output_directory, primers): + primer_sets = { + "ARTICv1": "https://raw.githubusercontent.com/artic-network/artic-ncov2019/master/primer_schemes/nCoV-2019/V1/nCoV-2019.bed", + "ARTICv2": "https://raw.githubusercontent.com/artic-network/artic-ncov2019/master/primer_schemes/nCoV-2019/V2/nCoV-2019.bed", + "ARTICv3": "https://raw.githubusercontent.com/artic-network/artic-ncov2019/master/primer_schemes/nCoV-2019/V3/nCoV-2019.bed", + } + + data = [] + for name, url in primer_sets.items(): + if name not in primers: + continue + response = requests.get(url) + if response.status_code != 200: + print( + "Error: download of", + url, + "failed with code", + response.status_code, + file=sys.stderr, + ) + exit(response.status_code) + bed_output_filename = os.path.join(output_directory, name + ".bed") + open(bed_output_filename, "w").write(response.text) + description = name[:-2] + " " + name[-2:] + " primer set" + data.append(dict(value=name, path=bed_output_filename, description=description)) + return data + + +def install_primer_file( + output_directory, input_filename, primer_name, primer_description +): + name = re.sub(r"\W", "", str(primer_name).replace(" ", "_")) + output_filename = os.path.join(output_directory, name + ".bed") + shutil.copyfile(input_filename, output_filename) + data = [dict(value=name, description=primer_description, path=output_filename)] + return data + + +class SplitArgs(argparse.Action): + def __call__(self, parser, namespace, values, option_string=None): + setattr(namespace, self.dest, values.split(",")) + + +if __name__ == "__main__": + parser = argparse.ArgumentParser( + description="Fetch ARTIC SARS-CoV-2 primer files for Galaxy/IRIDA use" + ) + parser.add_argument( + "--output_directory", default="tmp", help="Directory to write output to" + ) + primer_file = parser.add_argument_group() + primer_file.add_argument( + "--primer_file", help="BED format file containing primer scheme" + ) + primer_file.add_argument( + "--primer_name", + help="Name of primer scheme (one word). Required if --primer_file is used", + ) + primer_file.add_argument( + "--primer_description", + help="Description of primer scheme. Required if --primer_file is used", + ) + artic = parser.add_argument_group() + artic.add_argument( + "--artic_primers", + action=SplitArgs, + help="Comma separated list of primers to fetch", + ) + parser.add_argument( + "galaxy_datamanager_filename", + help="Galaxy JSON format file describing data manager inputs", + ) + args = parser.parse_args() + + if args.artic_primers is None and args.primer_file is None: + print( + "One of --artic_primers or --primer_file + --primer_name + --primer_description is required.", + file=sys.stderr, + ) + exit(1) + elif args.primer_file is not None and ( + args.primer_name is None or args.primer_description is None + ): + print( + "If --primer_file is used --primer_name and --primer_description is also required", + file=sys.stderr, + ) + exit(1) + elif args.primer_file is not None and args.artic_primers is not None: + print( + "Only one of --artic_primers or --primer_file + --primer_name + --primer_description can be chosen" + ) + exit(1) + + config = json.load(open(args.galaxy_datamanager_filename)) + output_directory = config.get("output_data", [{}])[0].get("extra_files_path", None) + if output_directory is None: + output_directory = args.output_directory + + if not os.path.isdir(output_directory): + os.makedirs(output_directory) + + data_manager_dict = {} + data_manager_dict["data_tables"] = json.load( + open(args.galaxy_datamanager_filename) + ).get("data_tables", {}) + data_manager_dict["data_tables"] = data_manager_dict.get("data_tables", {}) + data_manager_dict["data_tables"][DATA_TABLE_NAME] = data_manager_dict[ + "data_tables" + ].get(DATA_TABLE_NAME, []) + + if args.artic_primers: + data = fetch_artic_primers(output_directory, args.artic_primers) + else: + data = install_primer_file( + output_directory, + args.primer_file, + args.primer_name, + args.primer_description, + ) + + data_manager_dict["data_tables"][DATA_TABLE_NAME].extend(data) + print(data_manager_dict) + json.dump(data_manager_dict, open(args.galaxy_datamanager_filename, "w"))