Mercurial > repos > dfornika > data_manager_build_mash_sketch_refseq
changeset 0:9d53c3f65ee6 draft
planemo upload for repository https://github.com/dfornika/galaxytools/tree/master/data_managers/data_manager_build_mash_sketch_refseq/ commit b12e918e7ed6127297b258843f9c107c40878881-dirty
author | dfornika |
---|---|
date | Wed, 06 Mar 2019 18:52:43 -0500 |
parents | |
children | 0240f84d8a34 |
files | data_manager/mash_build_sketch_refseq.py data_manager/mash_build_sketch_refseq.xml data_manager_conf.xml tool-data/mash_sketches.loc.sample |
diffstat | 3 files changed, 148 insertions(+), 0 deletions(-) [+] |
line wrap: on
line diff
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/data_manager/mash_build_sketch_refseq.py Wed Mar 06 18:52:43 2019 -0500 @@ -0,0 +1,99 @@ +#!/usr/bin/env python + +from __future__ import print_function + +import argparse +import datetime +import errno +import json +import os +import string +import subprocess +import sys + + +DATA_TABLE_NAME = "mash_sketches" + +def run(args, cwd): + proc = subprocess.Popen(args=args, shell=False, cwd=cwd) + return_code = proc.wait() + if return_code: + print("Error building sketch.", file=sys.stderr) + sys.exit( return_code ) + +def mash_build_sketch_refseq(data_manager_dict, sketch_type, target_directory, data_table_name=DATA_TABLE_NAME): + + now = datetime.datetime.utcnow().strftime("%Y-%m-%dT%H%M%SZ") + + database_value = "_".join([ + now, + "refseq" + "." + + sketch_type + "." + + "k21.s1000.msh", + ]) + + database_name = " ".join([ + "refseq" + "." + + sketch_type + "." + + "k21.s1000.msh" + "(Created:", + now + ")" + ]) + + database_path = database_value + + args = [ + 'https://gembox.cbcb.umd.edu/mash/refseq.genomes.k21s1000.msh' + ] + + run(['wget'] + args, target_directory) + + data_table_entry = { + "value": database_value, + "name": database_name, + "path": database_path, + } + + _add_data_table_entry(data_manager_dict, data_table_entry) + + +def _add_data_table_entry(data_manager_dict, data_table_entry, data_table_name=DATA_TABLE_NAME): + 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, [] ) + data_manager_dict['data_tables'][data_table_name].append( data_table_entry ) + return data_manager_dict + + +def main(): + parser = argparse.ArgumentParser() + parser.add_argument('data_manager_json') + parser.add_argument('--sketch-type', dest='sketch_type', help='Sketch Type (genome or plasmid or genome+plasmid)' ) + parser.add_argument( '-t', '--threads', dest='threads', default=1, help='threads' ) + + args = parser.parse_args() + + data_manager_input = json.loads(open(args.data_manager_json).read()) + + target_directory = data_manager_input['output_data'][0]['extra_files_path'] + + try: + os.mkdir( target_directory ) + except OSError as exc: + if exc.errno == errno.EEXIST and os.path.isdir( target_directory ): + pass + else: + raise + + data_manager_output = {} + + mash_build_sketch_refseq( + data_manager_output, + args.sketch_type, + target_directory, + ) + + open(args.data_manager_json, 'wb').write(json.dumps(data_manager_output)) + + +if __name__ == "__main__": + main()
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/data_manager/mash_build_sketch_refseq.xml Wed Mar 06 18:52:43 2019 -0500 @@ -0,0 +1,31 @@ +<?xml version="1.0"?> +<tool id="mash_build_sketch_refseq" name="Mash RefSeq Sketch Builder" tool_type="manage_data" version="2.0"> + <description>database builder</description> + <requirements> + <requirement type="package" version="2.1">mash</requirement> + </requirements> + <version_command>mash --version</version_command> + <command detect_errors="exit_code"> + <![CDATA[ + python '$__tool_directory__/mash_build_sketch_refseq.py' + '${out_file}' + --threads \${GALAXY_SLOTS:-1} + --sketch-type ${sketch_type} + ]]> + </command> + <inputs> + <param name="sketch_type" type="select" multiple="false" label="Select MiniKraken2 database version to download"> + <option value="genome">genome</option> + <option value="plasmid">plasmid</option> + <option value="genome+plasmid">genome+plasmid</option> + </param> + </inputs> + <outputs> + <data name="out_file" format="data_manager_json" /> + </outputs> + <help> + </help> + <citations> + <citation type="doi">10.1186/gb-2014-15-3-r46</citation> + </citations> +</tool>
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/data_manager_conf.xml Wed Mar 06 18:52:43 2019 -0500 @@ -0,0 +1,18 @@ +<data_managers> + <data_manager tool_file="data_manager/mash_build_sketch_refseq.xml" id="mash_build_sketch_refseq" version="1.0"> + <data_table name="mash_sketches"> + <output> + <column name="value"/> + <column name="name"/> + <column name="path" output_ref="out_file"> + <move type="file"> + <source>${path}</source> + <target base="${GALAXY_DATA_MANAGER_DATA_PATH}">mash_sketches/${path}</target> + </move> + <value_translation>${GALAXY_DATA_MANAGER_DATA_PATH}/mash_sketches/${path}</value_translation> + <value_translation type="function">abspath</value_translation> + </column> + </output> + </data_table> + </data_manager> +</data_managers>