Mercurial > repos > dfornika > data_manager_mash_sketch_builder
changeset 1:af86557c1b87 draft
"planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/data_managers/data_manager_mash_sketch_builder/ commit 20558d9de7b6dd375a8835698f231cbc0c83446d-dirty"
author | dfornika |
---|---|
date | Sat, 22 Feb 2020 01:01:25 +0000 |
parents | 647eedfd1556 |
children | d892b30d5492 |
files | data_manager/mash_sketch_builder.py data_manager/mash_sketch_builder.xml macros.xml |
diffstat | 2 files changed, 22 insertions(+), 74 deletions(-) [+] |
line wrap: on
line diff
--- a/data_manager/mash_sketch_builder.py Sat Feb 22 00:30:35 2020 +0000 +++ b/data_manager/mash_sketch_builder.py Sat Feb 22 01:01:25 2020 +0000 @@ -19,22 +19,15 @@ sketch_path = os.path.join(UUID, "sketch"), args = [ - '--threads', str(kraken2_args["threads"]), '-k', str(mash_sketch_args["kmer_size"]), '-s', str(mash_sketch_args["sketch_size"]), - '-o', sketch_path + '-o', sketch_path, + '-p', str(mash_sketch_args["threads"]), + str(mash_sketch_args["fasta"]), ] subprocess.check_call(['mash', 'sketch'] + args, cwd=target_directory) - if kraken2_args["clean"]: - args = [ - '--threads', str(kraken2_args["threads"]), - '--clean', - '--db', database_path - ] - - subprocess.check_call(['kraken2-build'] + args, cwd=target_directory) data_table_entry = { 'data_tables': { @@ -56,6 +49,7 @@ parser.add_argument('data_manager_json') parser.add_argument('--kmer-size', dest='kmer_size', type=int, default=35, help='kmer length') parser.add_argument('--sketch-size', dest='sketch_size', type=int, default=31, help='minimizer length') + parser.add_argument('--fasta', dest='fasta', type='string', help='Fasta file to sketch') parser.add_argument('--threads', dest='threads', default=1, help='threads') parser.add_argument('--sketch-name', dest='sketch_name', help='Name for sketch') args = parser.parse_args() @@ -75,11 +69,18 @@ data_manager_output = {} mash_sketch_args = { - "kmer_size": args.kmer_len, - "sketch_size": args.minimizer_len, + "kmer_size": args.kmer_size, + "sketch_size": args.sketch_size, + "fasta": args.fasta, "threads": args.threads, } + data_manager_output = mash_sketch( + mash_sketch_args, + sketch_name, + target_directory, + ) + open(args.data_manager_json, 'w').write(json.dumps(data_manager_output, sort_keys=True))
--- a/data_manager/mash_sketch_builder.xml Sat Feb 22 00:30:35 2020 +0000 +++ b/data_manager/mash_sketch_builder.xml Sat Feb 22 01:01:25 2020 +0000 @@ -1,14 +1,6 @@ <?xml version="1.0"?> <tool id="mash_sketch_builder" name="Mash Sketch" tool_type="manage_data" version="2.1+galaxy0" profile="18.09"> - <macros> - <xml name="common_params"> - <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" /> - <param name="clean" type="boolean" truevalue="--clean" falsevalue="" checked="true" label="Clean up extra files" /> - </xml> - </macros> - <description>database builder</description> + <description>builder</description> <requirements> <requirement type="package" version="2.1">mash</requirement> <requirement type="package" version="3.7">python</requirement> @@ -18,68 +10,23 @@ <![CDATA[ python '$__tool_directory__/mash_sketch_builder.py' '${out_file}' - -s '${sketch_size}' - -k '${kmer_size}' - #if str ( $reads_assembly.reads_assembly_selector ) == "reads" - -m '${reads_assembly.minimum_kmer_copies}' - -r - #if str( $reads_assembly.reads_input.reads_input_selector ) == "paired" - '$reads_assembly.reads_input.reads_1' '$reads_assembly.reads_input.reads_2' - #end if - #if str( $reads_assembly.reads_input.reads_input_selector ) == "paired_collection" - '$reads_assembly.reads_input.reads.forward' '$reads_assembly.reads_input.reads.reverse' - #end if - #if str( $reads_assembly.reads_input.reads_input_selector ) == "single" - '$reads_assembly.reads_input.reads' - #end if - #elif str ( $reads_assembly.reads_assembly_selector ) == "assembly" - -p \${GALAXY_SLOTS:-1} - '${assembly}' - ${reads_assembly.individual_sequences} - #end if - -o 'sketch' + --fasta '${all_fasta_source.fields.path}' + --sketch-size '${sketch_size}' + --kmer-size '${kmer_size}' ]]> </command> <inputs> - <conditional name="reads_assembly"> - <param name="reads_assembly_selector" type="select" label="Input: Reads or Assemblies"> - <option selected="True" value="reads">Reads</option> - <option value="assembly">Assembly</option> - </param> - <when value="reads"> - <conditional name="reads_input"> - <param name="reads_input_selector" type="select" label="Single or Paired-end reads" help="Select between paired and single end data"> - <option value="paired">Paired</option> - <option value="single">Single</option> - <option value="paired_collection">Paired Collection</option> - </param> - <when value="paired"> - <param name="reads_1" type="data" format="@INTYPES@" label="Select first set of reads" help="Specify dataset with forward reads"/> - <param name="reads_2" type="data" format="@INTYPES@" label="Select second set of reads" help="Specify dataset with reverse reads"/> - </when> - <when value="single"> - <param name="reads" type="data" format="@INTYPES@" label="Select fastq dataset" help="Specify dataset with single reads"/> - </when> - <when value="paired_collection"> - <param name="reads" format="@INTYPES@" type="data_collection" collection_type="paired" label="Select a paired collection" help="See help section for an explanation of dataset collections"/> - </when> - </conditional> - <param type="integer" name="minimum_kmer_copies" argument="-m" value="1" min="1" max="1000" label="Minimum copies of each k-mer required to pass noise filter"/> - </when> - <when value="assembly"> - <param name="assembly" type="data" format="fasta,fasta.gz" label="Assembly"/> - <param type="boolean" name="individual_sequences" truevalue="-i" falsevalue="" label="Sketch individual Sequences"/> - </when> - </conditional> - <param type="integer" name="sketch_size" argument="-s" value="1000" min="10" max="1000000" label="Sketch size" /> - <param type="integer" name="kmer_size" argument="-k" value="21" min="1" max="32" label="kmer size" /> + <param name="all_fasta_source" type="select" label="Source FASTA Sequence"> + <options from_data_table="all_fasta"/> + </param> + <param type="integer" name="sketch_size" value="1000" min="10" max="1000000" label="Sketch size" /> + <param type="integer" name="kmer_size" value="21" min="1" max="32" label="kmer size" /> </inputs> <outputs> <data name="out_file" format="data_manager_json" /> </outputs> <tests> <test> - <param name="reads_assembly_selector" value="assembly" /> <param name="assembly" value="test_assembly.fasta"/> <param name="sketch_name" value="Test Sketch" /> <output name="out_file" value="mash_sketch_data_manager.json" />