Mercurial > repos > dfornika > mash
changeset 4:033da434cfaf draft
planemo upload for repository https://github.com/dfornika/galaxy/tree/master/tools/mash commit 54c5ec0b3958115ccb9071184268bfe4f7425baa-dirty
| author | dfornika |
|---|---|
| date | Mon, 28 Jan 2019 22:00:44 -0500 |
| parents | 351388dbbc0a |
| children | 4bfeab125a68 |
| files | mash_dist.xml mash_screen.xml test-data/mash_dist_NZ_MYON01000010.1_ERR024951_seqtk_sample_1000_1.tsv test-data/mash_dist_NZ_MYON01000010.1_ERR024951_seqtk_sample_1000_1and2.tsv test-data/mash_dist_NZ_MYON01000010.1_ERR024951_seqtk_sample_1000_2.tsv test-data/reads.fastq |
| diffstat | 6 files changed, 8139 insertions(+), 17 deletions(-) [+] |
line wrap: on
line diff
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/mash_dist.xml Mon Jan 28 22:00:44 2019 -0500 @@ -0,0 +1,108 @@ +<tool id="mash_dist" name="mash dist" version="@TOOL_VERSION@+galaxy0"> + <description>determines how well query sequences are contained within a pool of sequences.</description> + <macros> + <import>macros.xml</import> + </macros> + <requirements> + <requirement type="package" version="@TOOL_VERSION@">mash</requirement> + </requirements> + <version_command>mash --version</version_command> + <command detect_errors="exit_code"><![CDATA[ + #if str( $query_input.query_input_selector ) == "single" + #set $query $query_input.query.element_identifier + ln -s '$query_input.query' '$query' && + #end if + #if str( $query_input.query_input_selector ) == "paired" + #set $query $query_input.query_1.element_identifier + cat '$query_input.query_1' '$query_input.query_2' > '$query' && + #end if + #if str( $query_input.query_input_selector ) == "paired_collection" + #set $query $query_input.query.element_identifier + cat '$query_input.query.forward' '$query_input.query.reverse' > '$query' && + #end if + #if str( $query_input.query_input_selector ) == "collection" + cat '${ str($query_input.query).replace(',', '\' \'') }' > '$query' && + #end if + mash dist + -i $minimum_identity_to_report + -v $maximum_p_value_to_report + '$reference' + '$query' + > '$output' + ]]></command> + <inputs> + <param name="reference" type="data" format="@INTYPES@,binary" /> + <conditional name="query_input"> + <param name="query_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="collection">Collection</option> + <option value="paired_collection">Paired Collection</option> + </param> + <when value="paired"> + <param name="query_1" type="data" format="@INTYPES@" label="Select first set of reads" help="Specify dataset with forward reads"/> + <param name="query_2" type="data" format="@INTYPES@" label="Select second set of reads" help="Specify dataset with reverse reads"/> + </when> + <when value="single"> + <param name="query" type="data" format="@INTYPES@" label="Select fastq dataset" help="Specify dataset with single reads"/> + </when> + <when value="collection"> + <param name="query" format="@INTYPES@" type="data_collection" collection_type="single" label="Select a paired collection" help="See help section for an explanation of dataset collections"/> + </when> + <when value="paired_collection"> + <param name="pool" 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="float" name="minimum_identity_to_report" argument="-i" value="0." min="-1." max="1." /> + <param type="float" name="maximum_p_value_to_report" argument="-v" value="1." min="0." max="1."/> + </inputs> + <outputs> + <data name="output" format="tabular" /> + </outputs> + <tests> + <test> + <param name="reference" value="NZ_MYON01000010.1.msh"/> + <param name="query_input_selector" value="single"/> + <param name="query" value="ERR024951_seqtk_sample_1000_1.fastq"/> + <output name="output" file="mash_dist_NZ_MYON01000010.1_ERR024951_seqtk_sample_1000_1.tsv"/> + </test> + <test> + <param name="reference" value="NZ_MYON01000010.1.msh"/> + <param name="query_input_selector" value="single"/> + <param name="query" value="ERR024951_seqtk_sample_1000_2.fastq"/> + <output name="output" file="mash_dist_NZ_MYON01000010.1_ERR024951_seqtk_sample_1000_1.tsv"/> + </test> + <test> + <param name="reference" value="NZ_MYON01000010.1.msh"/> + <param name="query_input_selector" value="paired"/> + <param name="query_1" value="ERR024951_seqtk_sample_1000_1.fastq"/> + <param name="query_2" value="ERR024951_seqtk_sample_1000_2.fastq"/> + <output name="output" file="mash_dist_NZ_MYON01000010.1_ERR024951_seqtk_sample_1000_1and2.tsv"/> + </test> + </tests> + <help><![CDATA[ +Description: + + Estimate the distance of each query sequence to the reference. + Both the reference and queries can be fasta or fastq, gzipped or not, + or Mash sketch files (.msh) with matching k-mer sizes. Query files can also + be files of file names (see -l). Whole files are compared by default (see -i). + The output fields are [reference-ID, query-ID, distance, p-value, shared-hashes]. + ]]></help> + <citations> + <citation type="bibtex"> +@article{ondov2016mash, + title={Mash: fast genome and metagenome distance estimation using MinHash}, + author={Ondov, Brian D and Treangen, Todd J and Melsted, P{\'a}ll and Mallonee, Adam B and Bergman, Nicholas H and Koren, Sergey and Phillippy, Adam M}, + journal={Genome biology}, + volume={17}, + number={1}, + pages={132}, + year={2016}, + publisher={BioMed Central} + } + </citation> + </citations> +</tool>
--- a/mash_screen.xml Tue Jan 22 17:18:46 2019 -0500 +++ b/mash_screen.xml Mon Jan 28 22:00:44 2019 -0500 @@ -1,4 +1,4 @@ -<tool id="mash_screen" name="mash screen" version="@TOOL_VERSION@+galaxy0">fastq_ +<tool id="mash_screen" name="mash screen" version="@TOOL_VERSION@+galaxy0"> <description>determines how well query sequences are contained within a pool of sequences.</description> <macros> <import>macros.xml</import> @@ -26,22 +26,22 @@ > '$output' ]]></command> <inputs> - <conditional name="pool_input"> - <param name="pool_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="pool_1" type="data" format="@INTYPES@" label="Select first set of reads" help="Specify dataset with forward reads"/> - <param name="pool_2" type="data" format="@INTYPES@" label="Select second set of reads" help="Specify dataset with reverse reads"/> - </when> - <when value="single"> - <param name="pool" type="data" format="@INTYPES@" label="Select fastq dataset" help="Specify dataset with single reads"/> - </when> - <when value="paired_collection"> - <param name="pool" 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 name="pool_input"> + <param name="pool_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="pool_1" type="data" format="@INTYPES@" label="Select first set of reads" help="Specify dataset with forward reads"/> + <param name="pool_2" type="data" format="@INTYPES@" label="Select second set of reads" help="Specify dataset with reverse reads"/> + </when> + <when value="single"> + <param name="pool" type="data" format="@INTYPES@" label="Select fastq dataset" help="Specify dataset with single reads"/> + </when> + <when value="paired_collection"> + <param name="pool" 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 name="queries" type="data" format="binary" /> <param type="boolean" name="winner_takes_all" argument="-w" truevalue="-w" falsevalue=""/> @@ -71,6 +71,17 @@ <param name="pool_2" value="ERR024951_seqtk_sample_1000_2.fastq"/> <output name="output" file="mash_screen_NZ_MYON01000010.1_ERR024951_seqtk_sample_1000_1and2.tsv"/> </test> + <test> + <param name="queries" value="NZ_MYON01000010.1.msh"/> + <param name="pool_input_selector" value="paired_collection"/> + <param name="pool"> + <collection type="paired"> + <element name="forward" value="ERR024951_seqtk_sample_1000_1.fastq"/> + <element name="reverse" value="ERR024951_seqtk_sample_1000_2.fastq"/> + </collection> + </param> + <output name="output" file="mash_screen_NZ_MYON01000010.1_ERR024951_seqtk_sample_1000_1and2.tsv"/> + </test> </tests> <help><