Mercurial > repos > dfornika > data_manager_build_kraken2_database

changeset 0:7e43f66abe9d draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/dfornika/galaxytools/tree/master/data_managers/data_manager_build_kraken2_database/ commit cc42cd530378bfa3df83b08b6b41aaf2f5072b49-dirty
author dfornika
date Thu, 28 Feb 2019 02:32:38 -0500
parents
children 32d04371ed19
files data_manager/kraken2_build_standard.py data_manager/kraken2_build_standard.xml data_manager/kraken2_custom_database_builder.xml data_manager/kraken2_standard_database_builder.xml data_manager/make_json.py data_manager_conf.xml
diffstat 6 files changed, 288 insertions(+), 0 deletions(-) [+]
line wrap: on
line diff
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/data_manager/kraken2_build_standard.py	Thu Feb 28 02:32:38 2019 -0500
@@ -0,0 +1,96 @@
+#!/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
+
+from pprint import pprint
+
+DATA_TABLE_NAME = "kraken2_databases"
+
+def kraken2_build_standard(data_manager_dict, kraken2_args, params, target_directory, data_table_name=DATA_TABLE_NAME):
+    today = datetime.date.today().isoformat()
+    database_name = "_".join([today, "standard"])
+    args = [
+        'kraken2-build',
+        '--threads', str(kraken2_args["threads"]),
+        '--standard',
+        '--kmer-len', str(kraken2_args["kmer_len"]),
+        '--minimizer-len', str(kraken2_args["minimizer_len"]),
+        '--minimizer-spaces', str(kraken2_args["minimizer_spaces"]),
+        '--db', database_name
+    ]
+    pprint(args)
+    proc = subprocess.Popen(args=args, shell=False, cwd=target_directory)
+    return_code = proc.wait()
+    if return_code:
+        print("Error building database.", file=sys.stderr)
+        sys.exit( return_code )
+
+    data_table_entry = {
+        "value": database_name,
+        "name": database_name,
+        "path": database_name
+    }
+    
+    _add_data_table_entry(data_manager_dict, data_table_name, data_table_entry)
+
+
+def _add_data_table_entry(data_manager_dict, data_table_name, data_table_entry):
+    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('params')
+    parser.add_argument( '-k', '--kmer-len', dest='kmer_len', type=int, default=35, help='kmer length' )
+    parser.add_argument( '-m', '--minimizer-len', dest='minimizer_len', type=int, default=31, help='minimizer length' )
+    parser.add_argument( '-s', '--minimizer-spaces', dest='minimizer_spaces', default=6, help='minimizer spaces' )
+    parser.add_argument( '-t', '--threads', dest='threads', default=1, help='threads' )
+    args = parser.parse_args()
+
+    kraken2_args = {
+        "kmer_len": args.kmer_len,
+        "minimizer_len": args.minimizer_len,
+        "minimizer_spaces": args.minimizer_spaces,
+        "threads": args.threads,
+    }
+    
+    params = json.loads(open(args.params).read())
+    pprint(params)
+    target_directory = params['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_dict = {}
+
+    # build the index
+    kraken2_build_standard(
+        data_manager_dict,
+        kraken2_args,
+        params,
+        target_directory
+    )
+
+    # save info to json file
+    open(args.params, 'wb').write(json.dumps(data_manager_dict))
+
+
+if __name__ == "__main__":
+    main()
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/data_manager/kraken2_build_standard.xml	Thu Feb 28 02:32:38 2019 -0500
@@ -0,0 +1,31 @@
+<?xml version="1.0"?>
+<tool id="kraken2_build_standard" name="Kraken2 Standard Database Builder (subprocess)" tool_type="manage_data" version="2.0.7_beta">
+    <description>database builder</description>
+    <requirements>
+        <requirement type="package" version="2.0.7_beta">kraken2</requirement>
+    </requirements>
+    <version_command>kraken2 -version | head -n 1 | awk '{print $NF}'</version_command>
+    <command detect_errors="exit_code">
+    <![CDATA[
+        python '$__tool_directory__/kraken2_build_standard.py'
+          '${out_file}'
+          --threads \${GALAXY_SLOTS:-1}
+          --kmer-len ${kmer_len}
+          --minimizer-len ${minimizer_len}
+          --minimizer-spaces ${minimizer_spaces}
+    ]]>
+    </command>
+    <inputs>
+        <param name="kmer_len" type="integer" value="35" label="K-mer length in BP" />
+        <param name="minimizer_len" type="integer" value="31" label="Minimizer length" />
+	<param name="minimizer_spaces" type="integer" value="6" label="Minimizer spaces" />
+    </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/kraken2_custom_database_builder.xml	Thu Feb 28 02:32:38 2019 -0500
@@ -0,0 +1,50 @@
+<?xml version="1.0"?>
+<tool id="kraken2_custom_database_builder" name="Kraken2 Custom Database Builder" tool_type="manage_data" version="2.0.7_beta">
+    <description>database builder</description>
+    <requirements>
+        <requirement type="package" version="2.0.7_beta">kraken2</requirement>
+    </requirements>
+    <version_command>kraken2 -version | head -n 1 | awk '{print $NF}'</version_command>
+    <command detect_errors="exit_code">
+    <![CDATA[
+        #set $library_list = str($libraries).split(',')
+        mkdir kraken2-database && cd kraken2-database &&
+        kraken2-build --threads \${GALAXY_SLOTS:-1} --download-taxonomy --db '${database_name}' &&
+        #for $library in $library_list:
+            #if str($library) != 'None':
+                kraken2-build --threads \${GALAXY_SLOTS:-1} --download-library $library --db '${database_name}' &&
+            #end if
+        #end for
+        #for $fasta in $input_fasta:
+            #if str($fasta) != 'None':
+                kraken2-build --threads \${GALAXY_SLOTS:-1} --db '${database_name}' --add-to-library '$fasta' &&
+            #end if
+        #end for
+        kraken2-build --threads \${GALAXY_SLOTS:-1} --db '${database_name}' --build --kmer-len ${kmer_len} --minimizer-len ${minimizer_len} &&
+        cd .. &&
+        python '$__tool_directory__/make_json.py' --db '${database_name}' --out '${out_file}'
+    ]]>
+    </command>
+    <inputs>
+        <param name="database_name" type="text" label="Name for this database" />
+        <param name="libraries" type="select" multiple="true" label="Select partial library to download">
+            <option value="archaea">Archaea</option>
+	    <option value="bacteria">Bacteria</option>
+            <option value="plasmid">Plasmid</option>
+            <option value="viral">Viral</option>
+            <option value="human">Human</option>
+        </param>
+        <param name="input_fasta" type="data" format="fasta" multiple="True" optional="true" label="Select history item" />
+        <param name="kmer_len" type="integer" value="35" label="K-mer length in BP" />
+        <param name="minimizer_len" type="integer" value="31" label="Minimizer length" />
+	<param name="minimizer_spaces" type="integer" value="6" label="Minimizer spaces" />
+    </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/kraken2_standard_database_builder.xml	Thu Feb 28 02:32:38 2019 -0500
@@ -0,0 +1,37 @@
+<?xml version="1.0"?>
+<tool id="kraken2_standard_database_builder" name="Kraken2 Standard Database Builder" tool_type="manage_data" version="2.0.7_beta">
+    <description>database builder</description>
+    <requirements>
+        <requirement type="package" version="2.0.7_beta">kraken2</requirement>
+    </requirements>
+    <version_command>kraken2 -version | head -n 1 | awk '{print $NF}'</version_command>
+    <command detect_errors="exit_code">
+    <![CDATA[
+        #import datetime
+	#set kraken2_db_name = datetime.date.today().strftime('%Y-%m-%d') + "_standard"
+        mkdir kraken2-database && cd kraken2-database &&
+        kraken2-build 
+          --standard
+	  --threads \${GALAXY_SLOTS:-1}
+          --db '${kraken2_db_name}'
+          --kmer-len ${kmer_len}
+          --minimizer-len ${minimizer_len}
+          --minimizer-spaces ${minimizer_spaces} &&
+        cd .. &&
+        python '$__tool_directory__/make_json.py' --db '${kraken2_db_name}' --out '${out_file}'
+    ]]>
+    </command>
+    <inputs>
+        <param name="kmer_len" type="integer" value="35" label="K-mer length in BP" />
+        <param name="minimizer_len" type="integer" value="31" label="Minimizer length" />
+	<param name="minimizer_spaces" type="integer" value="6" label="Minimizer spaces" />
+    </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/make_json.py	Thu Feb 28 02:32:38 2019 -0500
@@ -0,0 +1,27 @@
+import argparse
+import json
+import os
+import shutil
+
+
+def main(args):
+    data_manager_entry = {}
+    data_manager_entry['value'] = args.database.lower()
+    data_manager_entry['name'] = args.database
+    data_manager_entry['path'] = '.'
+    data_manager_json = dict(data_tables=dict(kraken2_databases=data_manager_entry))
+    params = json.loads(open(args.output).read())
+    target_directory = params['output_data'][0]['extra_files_path']
+    os.mkdir(target_directory)
+    output_path = os.path.join(os.getcwd(), 'kraken2-database')
+    for filename in os.listdir(output_path):
+        shutil.move(os.path.join(output_path, filename), target_directory)
+    file(args.output, 'w').write(json.dumps(data_manager_json))
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(description='Create data manager json.')
+    parser.add_argument('--db', dest='database', action='store', help='Database name')
+    parser.add_argument('--out', dest='output', action='store', help='JSON filename')
+    args = parser.parse_args()
+    main(args)
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/data_manager_conf.xml	Thu Feb 28 02:32:38 2019 -0500
@@ -0,0 +1,47 @@
+<data_managers>
+    <data_manager tool_file="data_manager/kraken2_custom_database_builder.xml" id="kraken2_custom_database_builder" version="1.0">
+	<data_table name="kraken2_databases">
+	    <output>
+		<column name="value"/>
+		<column name="name"/>
+		<column name="path" output_ref="out_file">
+		    <move type="directory" relativize_symlinks="True">
+			<target base="${GALAXY_DATA_MANAGER_DATA_PATH}">kraken2_database/${value}</target>
+		    </move>
+		    <value_translation>${GALAXY_DATA_MANAGER_DATA_PATH}/kraken2_database/${value}</value_translation>
+		    <value_translation type="function">abspath</value_translation>
+		</column>
+	    </output>
+	</data_table>
+    </data_manager>
+    <data_manager tool_file="data_manager/kraken2_standard_database_builder.xml" id="kraken2_standard_database_builder" version="1.0">
+	<data_table name="kraken2_databases">
+	    <output>
+		<column name="value"/>
+		<column name="name"/>
+		<column name="path" output_ref="out_file">
+		    <move type="directory" relativize_symlinks="True">
+			<target base="${GALAXY_DATA_MANAGER_DATA_PATH}">kraken2_database/${value}</target>
+		    </move>
+		    <value_translation>${GALAXY_DATA_MANAGER_DATA_PATH}/kraken2_database/${value}</value_translation>
+		    <value_translation type="function">abspath</value_translation>
+		</column>
+	    </output>
+	</data_table>
+    </data_manager>
+    <data_manager tool_file="data_manager/kraken2_build_standard.xml" id="kraken2_build_standard" version="1.0">
+	<data_table name="kraken2_databases">
+	    <output>
+		<column name="value"/>
+		<column name="name"/>
+		<column name="path" output_ref="out_file">
+		    <move type="directory" relativize_symlinks="True">
+			<target base="${GALAXY_DATA_MANAGER_DATA_PATH}">kraken2_database/${value}</target>
+		    </move>
+		    <value_translation>${GALAXY_DATA_MANAGER_DATA_PATH}/kraken2_database/${value}</value_translation>
+		    <value_translation type="function">abspath</value_translation>
+		</column>
+	    </output>
+	</data_table>
+    </data_manager>
+</data_managers>