Mercurial > repos > devteam > data_manager_sam_fasta_index_builder
changeset 11:defd3e9ddd2d draft default tip
planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/data_managers/data_manager_sam_fasta_index_builder commit 57d05675396f0b44265fb4dbc8f9f891c6073219
| author | iuc |
|---|---|
| date | Thu, 05 Dec 2024 06:49:31 +0000 |
| parents | a256278e5bff |
| children | |
| files | data_manager/data_manager_sam_fasta_index_builder.py data_manager/data_manager_sam_fasta_index_builder.xml data_manager_conf.xml test-data/all_fasta.loc test-data/sam_fasta_data_manager.1.json test-data/sam_fasta_data_manager.2.json test-data/sam_fasta_data_manager.json tool_data_table_conf.xml.sample |
| diffstat | 8 files changed, 66 insertions(+), 119 deletions(-) [+] |
line wrap: on
line diff
--- a/data_manager/data_manager_sam_fasta_index_builder.py Mon Dec 18 09:37:01 2023 +0000 +++ /dev/null Thu Jan 01 00:00:00 1970 +0000 @@ -1,102 +0,0 @@ -#!/usr/bin/env python -# Dan Blankenberg - -import json -import optparse -import os -import subprocess -import sys -import tempfile - -CHUNK_SIZE = 2**20 - -DEFAULT_DATA_TABLE_NAME = "fasta_indexes" - - -def get_id_name(params, dbkey, fasta_description=None): - # TODO: ensure sequence_id is unique and does not already appear in location file - sequence_id = params['param_dict']['sequence_id'] - if not sequence_id: - sequence_id = dbkey - - sequence_name = params['param_dict']['sequence_name'] - if not sequence_name: - sequence_name = fasta_description - if not sequence_name: - sequence_name = dbkey - return sequence_id, sequence_name - - -def build_sam_index(data_manager_dict, fasta_filename, target_directory, dbkey, sequence_id, sequence_name, data_table_name=DEFAULT_DATA_TABLE_NAME): - # TODO: allow multiple FASTA input files - assert os.path.exists(fasta_filename), 'FASTA file "%s" is missing, cannot build samtools index.' % fasta_filename - fasta_base_name = os.path.split(fasta_filename)[-1] - sym_linked_fasta_filename = os.path.join(target_directory, fasta_base_name) - os.symlink(fasta_filename, sym_linked_fasta_filename) - - args = ['samtools', 'faidx'] - args.append(sym_linked_fasta_filename) - tmp_stderr = tempfile.NamedTemporaryFile(prefix="tmp-data-manager-sam_fa_index_builder-stderr") - proc = subprocess.Popen(args=args, shell=False, cwd=target_directory, stderr=tmp_stderr.fileno()) - return_code = proc.wait() - if return_code: - tmp_stderr.flush() - tmp_stderr.seek(0) - sys.stderr.write("Error building index:\n") - while True: - chunk = tmp_stderr.read(CHUNK_SIZE) - if not chunk: - break - sys.stderr.write(chunk) - sys.exit(return_code) - tmp_stderr.close() - data_table_entry = dict(value=sequence_id, dbkey=dbkey, name=sequence_name, path=fasta_base_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 = optparse.OptionParser() - parser.add_option('-f', '--fasta_filename', dest='fasta_filename', action='store', type="string", default=None, help='fasta_filename') - parser.add_option('-d', '--fasta_dbkey', dest='fasta_dbkey', action='store', type="string", default=None, help='fasta_dbkey') - parser.add_option('-t', '--fasta_description', dest='fasta_description', action='store', type="string", default=None, help='fasta_description') - parser.add_option('-n', '--data_table_name', dest='data_table_name', action='store', type="string", default=None, help='data_table_name') - (options, args) = parser.parse_args() - - filename = args[0] - - with open(filename) as fh: - params = json.load(fh) - target_directory = params['output_data'][0]['extra_files_path'] - os.mkdir(target_directory) - data_manager_dict = {} - - if options.fasta_dbkey in [None, '', '?']: - raise Exception('"%s" is not a valid dbkey. You must specify a valid dbkey.' % (options.fasta_dbkey)) - - sequence_id, sequence_name = get_id_name(params, dbkey=options.fasta_dbkey, fasta_description=options.fasta_description) - - # build the index - build_sam_index( - data_manager_dict, - options.fasta_filename, - target_directory, - options.fasta_dbkey, - sequence_id, - sequence_name, - data_table_name=options.data_table_name or DEFAULT_DATA_TABLE_NAME - ) - - # save info to json file - with open(filename, 'w') as fh: - json.dump(data_manager_dict, fh, sort_keys=True) - - -if __name__ == "__main__": - main()
--- a/data_manager/data_manager_sam_fasta_index_builder.xml Mon Dec 18 09:37:01 2023 +0000 +++ b/data_manager/data_manager_sam_fasta_index_builder.xml Thu Dec 05 06:49:31 2024 +0000 @@ -1,17 +1,39 @@ -<tool id="sam_fasta_index_builder" name="SAM FASTA index" tool_type="manage_data" version="0.0.6" profile="19.05"> +<tool id="sam_fasta_index_builder" name="SAM FASTA index" tool_type="manage_data" version="@WRAPPER_VERSION@+galaxy@VERSION_SUFFIX@" profile="23.0"> <description>builder</description> + <macros> + <token name="@WRAPPER_VERSION@">1.21</token> + <token name="@VERSION_SUFFIX@">0</token> + </macros> <requirements> - <requirement type="package" version="1.14">samtools</requirement> - <requirement type="package" version="3.7">python</requirement> + <requirement type="package" version="1.21">samtools</requirement> </requirements> <command detect_errors="exit_code"><![CDATA[ - python '$__tool_directory__/data_manager_sam_fasta_index_builder.py' - '${out_file}' - --fasta_filename '${all_fasta_source.fields.path}' - --fasta_dbkey '${all_fasta_source.fields.dbkey}' - --fasta_description '${all_fasta_source.fields.name}' - --data_table_name fasta_indexes + #set $fasta_file_name = str($all_fasta_source.fields.path).split('/')[-1] + mkdir -p '${out_file.extra_files_path}' && + ln -s '${all_fasta_source.fields.path}' '${out_file.extra_files_path}/${fasta_file_name}' && + cd '${out_file.extra_files_path}' && + samtools faidx '${out_file.extra_files_path}/${fasta_file_name}' && + cp '$dmjson' '$out_file' ]]></command> + <configfiles> + <configfile name="dmjson"><![CDATA[#slurp +#set $fasta_file_name = str($all_fasta_source.fields.path).split('/')[-1] +#set $value = $sequence_id or $all_fasta_source.fields.dbkey +#set $name = $sequence_name or $all_fasta_source.fields.name +{ + "data_tables":{ + "fasta_indexes":[ + { + "value": "${value}", + "dbkey": "${all_fasta_source.fields.dbkey}", + "name": "${name}", + "path": "${fasta_file_name}" + } + ] + } +} +]]></configfile> + </configfiles> <inputs> <param name="all_fasta_source" type="select" label="Source FASTA Sequence"> <options from_data_table="all_fasta"/> @@ -25,7 +47,13 @@ <tests> <test> <param name="all_fasta_source" value="phiX174"/> - <output name="out_file" value="sam_fasta_data_manager.json"/> + <output name="out_file" value="sam_fasta_data_manager.1.json"/> + </test> + <test> + <param name="all_fasta_source" value="phiX174"/> + <param name="sequence_name" value="Galeocerdo cuvier"/> + <param name="sequence_id" value="tigHai1"/> + <output name="out_file" value="sam_fasta_data_manager.2.json"/> </test> </tests>
--- a/data_manager_conf.xml Mon Dec 18 09:37:01 2023 +0000 +++ b/data_manager_conf.xml Thu Dec 05 06:49:31 2024 +0000 @@ -1,6 +1,5 @@ <?xml version="1.0"?> <data_managers> - <data_manager tool_file="data_manager/data_manager_sam_fasta_index_builder.xml" id="sam_fasta_index_builder"> <data_table name="fasta_indexes"> <output> @@ -10,13 +9,12 @@ <column name="path" output_ref="out_file" > <move type="directory" relativize_symlinks="True"> <!-- <source>${path}</source>--> <!-- out_file.extra_files_path is used as base by default --> <!-- if no source, eg for type=directory, then refers to base --> - <target base="${GALAXY_DATA_MANAGER_DATA_PATH}">${dbkey}/sam_indexes/${value}</target> + <target base="${GALAXY_DATA_MANAGER_DATA_PATH}">genomes/${dbkey}/sam_fasta_index/v1/${value}</target> </move> - <value_translation>${GALAXY_DATA_MANAGER_DATA_PATH}/${dbkey}/sam_indexes/${value}/${path}</value_translation> + <value_translation>${GALAXY_DATA_MANAGER_DATA_PATH}/genomes/${dbkey}/sam_fasta_index/v1/${value}/${path}</value_translation> <value_translation type="function">abspath</value_translation> </column> </output> </data_table> </data_manager> - </data_managers>
--- a/test-data/all_fasta.loc Mon Dec 18 09:37:01 2023 +0000 +++ b/test-data/all_fasta.loc Thu Dec 05 06:49:31 2024 +0000 @@ -16,4 +16,4 @@ #fasta file. So there will be multiple fasta files for each build, #such as with hg19 above. # -phiX174 phiX174 phiX174 ${__HERE__}/phiX174.fasta +phiX174 phiX174 phiX 174 ${__HERE__}/phiX174.fasta
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/test-data/sam_fasta_data_manager.1.json Thu Dec 05 06:49:31 2024 +0000 @@ -0,0 +1,12 @@ +{ + "data_tables":{ + "fasta_indexes":[ + { + "value": "phiX174", + "dbkey": "phiX174", + "name": "phiX 174", + "path": "phiX174.fasta" + } + ] + } +}
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/test-data/sam_fasta_data_manager.2.json Thu Dec 05 06:49:31 2024 +0000 @@ -0,0 +1,12 @@ +{ + "data_tables":{ + "fasta_indexes":[ + { + "value": "tigHai1", + "dbkey": "phiX174", + "name": "Galeocerdo cuvier", + "path": "phiX174.fasta" + } + ] + } +}
--- a/test-data/sam_fasta_data_manager.json Mon Dec 18 09:37:01 2023 +0000 +++ /dev/null Thu Jan 01 00:00:00 1970 +0000 @@ -1,1 +0,0 @@ -{"data_tables": {"fasta_indexes": [{"dbkey": "phiX174", "name": "phiX174", "path": "phiX174.fasta", "value": "phiX174"}]}} \ No newline at end of file
--- a/tool_data_table_conf.xml.sample Mon Dec 18 09:37:01 2023 +0000 +++ b/tool_data_table_conf.xml.sample Thu Dec 05 06:49:31 2024 +0000 @@ -1,10 +1,10 @@ -<!-- Use the file tool_data_table_conf.xml.oldlocstyle if you don't want to update your loc files as changed in revision 4550:535d276c92bc--> <tables> <!-- Locations of all fasta files under genome directory --> <table name="all_fasta" comment_char="#"> <columns>value, dbkey, name, path</columns> <file path="tool-data/all_fasta.loc" /> </table> + <!-- Locations of fasta indexes generated by samtools faidx --> <table name="fasta_indexes" comment_char="#"> <columns>value, dbkey, name, path</columns> <file path="tool-data/fasta_indexes.loc" />