comparison instrain_compare.xml @ 0:dff92aac9f75 draft

"planemo upload for repository https://github.com/MrOlm/inStrain commit e6eae71231e551c08aa96afc9f15b8ba87676101"

0:dff92aac9f75

<tool id="instrain_compare" name="InStrain Compare" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@PROFILE@">
    <description>Compares multiple inStrain profiles (popANI, coverage_overlap, etc.) </description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <expand macro="edam_ontology"/>
    <expand macro="requirements"/>
    <version_command>inStrain compare --version</version_command>
    <command detect_errors="exit_code"><![CDATA[
#if $stb
    ln -s '$stb' 'stb_file.stb' &&
#end if
#if $other.genome
    ln -s '$other.genome' 'genome_file.stb' &&
#end if
#for $i, $s in enumerate($input_is)
    #if $s
        input_count=$i    
        mkdir -p $i-input.IS &&
        unzip '$s' -d '$i-input.IS/' &&
    #end if
#end for
inStrain compare
    --input
    #for $i, $s in enumerate($input_is)
        #if $s
            '$i-input.IS'
        #end if
    #end for    
    --output 'output.IS.COMPARE'
    --processes "\${GALAXY_SLOTS:-6}"
#if $stb
    --stb 'stb_file.stb'
#end if    
    --min_cov $variant_calling.min_cov
    --min_freq $variant_calling.min_freq
    --fdr $variant_calling.fdr
    $database.database_mode
    --breadth $database.breadth
#if $other.scaffolds
    --scaffolds '$other.scaffolds'
#end if
#if $other.genome
    --genome 'genome_file.stb'
#end if
    $other.store_coverage_overlap
    $other.store_mismatch_locations
    $other.include_self_comparisons
    $other.skip_plot_generation
    --group_length $other.group_length
    --ani_threshold $genome_clustering.ani_threshold
    --coverage_treshold $genome_clustering.coverage_treshold
    --clusterAlg '$genome_clustering.clusterAlg'
    ]]></command>    
    <inputs>
        <param name="input_is" type="data" format="zip" multiple="true" label="inStrain Profile IS folder" help=" The Zip files for the IS profiles outputs you want to compare"/>                    
        <param argument="--stb" type="data" format="tabular" optional="true" label="Scaffold to bin" help="This can be a file with each line listing a scaffold and a bin name, tab-seperated. This can also be a space-seperated list of .fasta files, with one genome per .fasta file. If nothing is provided, all scaffolds will be treated as belonging to the same genome"/>
        <section name="variant_calling" title="Variant Calling Options" expanded="true">
            <param argument="--min_cov" type="integer" value="5" label=" Minimum coverage to call an variant"/>
            <param argument="--min_freq" type="float" value="0.05" label="Minimum SNP frequency to confirm a SNV" help="Both this AND the FDR snp count cutoff must be true to call a SNP."/>
            <param argument="--fdr" type="float" value="1e-06" min="0" max="1" help="SNP false discovery rate- based on simulation data with a 0.1 percent error rate (Q30)"/>
        </section>
        <section name="database" title="Database Mode Parameters" expanded="true">
            <param argument="--database_mode" type="boolean" truevalue="--debugdatabase_mode" falsevalue="" checked="false" label="Automatically determine which genomes are present in each Profile and only compare scaffolds from those genomes." help="All profiles must have run Profile with the same .stb"/>
            <param argument="--breadth" type="float" value="0.5" label="Minimum breadth_minCov required to count a genome present"/>
        </section>
        <section name="other" title="Other Options" expanded="true">
            <param argument="--scaffolds" type="data" format="fasta" optional="true" label="Location to a list of scaffolds to compare. You can also make this a .fasta file and it will load the scaffold names"/>
            <param argument="--genome" type="data" format="tabular" optional="true" label="Run scaffolds belonging to this single genome only. Must provide an .stb file"/>
            <param argument="--store_coverage_overlap" type="boolean" truevalue="--store_coverage_overlap" falsevalue="" checked="false" label="Store coverage overlap on an mm level"/>
            <param argument="--store_mismatch_locations" type="boolean" truevalue="--store_mismatch_locations" falsevalue="" checked="false" label="Store the locations of SNPs"/>
            <param argument="--include_self_comparisons" type="boolean" truevalue="--include_self_comparisons" falsevalue="" checked="false" label="Compare IS profiles against themself"/>
            <param argument="--skip_plot_generation" type="boolean" truevalue="--skip_plot_generation" falsevalue="" checked="false" label="Dont create plots at the end of the run"/>
            <param argument="--group_length" type="integer" value="10000000" label="How many bp to compare simultaneously" help="higher will use more RAM and run more quickly"/>
        </section>
        <section name="genome_clustering" title="Genome Clustering Options" expanded="true">
            <param argument="--ani_threshold" type="float" value="0.99999" label="popANI threshold to cluster genomes at" help="Must provide .stb file to do so"/>
            <param argument="--coverage_treshold" type="float" value="0.1" label="Minimum percent_genome_compared for a genome comparison to count" help="if below the popANI will be set to 0"/>
            <param argument="--clusterAlg" type="select" label="Algorithm used to cluster genomes">
                <option value="average" selected="true">Average</option>
                <option value="single">Single</option>
                <option value="ward">Ward</option>
                <option value="complete">complete</option>
                <option value="centroid">centroid</option>
                <option value="weighted">weighted</option>
                <option value="median">median</option>
            </param>
        </section>
    </inputs>
    <outputs>
        <data name="comparisonsTable" format="tabular" from_work_dir="output.IS.COMPARE/output/output.IS.COMPARE_comparisonsTable.tsv" label="Comparisons Table: Summarizes the differences between two inStrain profiles on a scaffold by scaffold level" />
        <data name="pairwise_SNP_locations" format="tabular" from_work_dir="output.IS.COMPARE/output/output.IS.COMPARE_pairwise_SNP_locations.tsv" label="Pairwise SNP locations: Lists the locations of all differences between profiles." />
        <data name="genomeWide_compare" format="tabular" from_work_dir="output.IS.COMPARE/output/output.IS.COMPARE_genomeWide_compare.tsv" label="Genome Wide compare: A genome-level summary of the differences detected by inStrain compare." />
        <data format="tabular" name="strain_clusters" from_work_dir="output.IS.COMPARE/output/output.IS.COMPARE_strain_clusters.tsv" label="Strain clusters: Generate strain-level clusters" />
        <data format="pdf" name="inStrainCompare_dendrograms" from_work_dir="output.IS.COMPARE/figures/output.IS.COMPARE_inStrainCompare_dendrograms.pdf" label="inStrain Compare dendrograms: genomeWide microdiveristy metrics" />       
    </outputs>
    <tests>        
    <test expect_num_outputs="5">
        <param name="stb" value="N5_271_010G1.maxbin2.stb"/>
        <param name="input_is" value="N5_271_010G1_scaffold_min1000.fa-vs-N5_271_010G1.IS.zip,N5_271_010G1_scaffold_min1000.fa-vs-N5_271_010G2.IS.zip"/>
        <section name="variant_calling">
            <param name="min_cov" value="5"/>
            <param name="min_freq" value="0.05"/>
            <param name="fdr" value="1e-06"/>
        </section>
        <section name="database">
            <param name="database_mode" value="false"/>
            <param name="breadth" value="0.5"/>
        </section>
        <section name="other">
            <param name="store_coverage_overlap" value="false"/>
            <param name="store_mismatch_locations" value="false"/>
            <param name="include_self_comparisons" value="false"/>
            <param name="skip_plot_generation" value="false"/>
            <param name="group_length" value="10000000"/>
        </section>
        <section name="genome_clustering">
            <param name="ani_threshold" value="0.99999"/>
            <param name="coverage_treshold" value="0.1"/>
            <param name="clusterAlg" value="average"/>
        </section>
        <output name="comparisonsTable">
            <assert_contents>
                <has_text text="N5_271_010G1_scaffold_73"/>
                <has_n_lines n="168"/>
                <has_n_columns n="11"/>
            </assert_contents>
        </output>
        <output name="pairwise_SNP_locations">
            <assert_contents>
                <has_n_lines n="0"/>
            </assert_contents>
        </output>
        <output name="genomeWide_compare">
            <assert_contents>
                <has_text text="name1"/>
                <has_n_lines n="3"/>
                <has_n_columns n="10"/>
            </assert_contents>
        </output>
        <output name="strain_clusters">
            <assert_contents>
                <has_text text="1_1"/>
                <has_n_lines n="5"/>
                <has_n_columns n="3"/>
            </assert_contents>
        </output>
        <output name="inStrainCompare_dendrograms">
            <assert_contents>
                <has_size value="384512" delta="10000" />
            </assert_contents>
        </output>
    </test>
    </tests>
    <help><![CDATA[
@HELP_HEADER@

Compare
=======

is part of the inStrain module that provides the ability to compare multiple inStrain profiles (created by running inStrain profile).

Note
====

inStrain can only compare inStrain profiles that have been mapped to the same .fasta file

inStrain compare does pairwise comparisons between each input inStrain profile. For each pair, a series of steps are undertaken:

1. All positions in which both IS_profile objects have at least min_cov coverage (5x by default) are identified. This information can be stored in the output by using the flag –store_coverage_overlap, but due to it’s size, it’s not stored by default.


2. Each position identified in step 1 is compared to calculate both conANI and popANI. The way that it compares positions is by testing whether the consensus base in sample 1 is detected at all in sample 2 and vice-versa. Detection of an allele in a sample is based on that allele being above the set -min_freq and -fdr. All detected differences between each pair of samples can be reported if the flag –store_mismatch_locations is set.


3. The coverage overlap and the average nucleotide identity for each scaffold is reported. For details on how this is done.


Inputs
======

Multiple inStrain profiles IS outputs (zip files), all mapped to the same .fasta file


Outputs
=======

1. comparisonsTable.tsv
   
   Summarizes the differences between two inStrain profiles on a scaffold by scaffold level

2. pairwise_SNP_locations.tsv

   Lists the locations of all differences between profiles. Because it’s a big file, this will only be created is you include the flag --store_mismatch_locations in your inStrain compare command.

3. genomeWide_compare.tsv

   A genome-level summary of the differences detected by inStrain compare. Generated by running inStrain genome_wide on the results of inStrain compare

4. strain_clusters.tsv

   The result of clustering the pairwise comparison data provided in genomeWide_compare.tsv to generate strain-level clusters. Performed using hierarchical clustering in the same manner as the program dRep

5. Compare dendrograms (PDF) figure/plot
   
   A dendrogram comparing all samples based on popANI and based on shared_bases.

    ]]></help> 
    <citations>
        <citation type="doi">10.1101/2020.01.22.915579</citation>
    </citations>
</tool>