diff stacks_gstacks.xml @ 0:d35cb34f2b85 draft

planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/stacks2 commit b395fa36fa826e26085820ba3a9faacaeddcb460
author iuc
date Mon, 01 Jul 2019 10:59:14 -0400
parents
children 27359c6bf3e3
line wrap: on
line diff
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/stacks_gstacks.xml	Mon Jul 01 10:59:14 2019 -0400
@@ -0,0 +1,382 @@
+<tool id="stacks2_gstacks" name="Stacks2: gstacks" profile="@PROFILE@" version="@STACKS_VERSION@+galaxy@WRAPPER_VERSION@">
+    <description>Call variants, genotypes and haplotype</description>
+    <macros>
+        <import>macros.xml</import>
+    </macros>
+    <expand macro="requirements">
+        <requirement type="package" version="1.9">samtools</requirement>
+    </expand>
+    <expand macro="version_cmd"/>
+    <command detect_errors="aggressive"><![CDATA[
+@FASTQ_INPUT_FUNCTIONS@
+
+mkdir bam_inputs stacks_outputs &&
+#if $mode_cond.mode_select == "denovo" and not $popmap:
+    ## since collections have no len .. yet
+    #try:
+        #set count = len($input_bam)
+    #except:
+        #set count = len($input_bam.keys())
+    #end try
+    #if count == 1:
+        #for $bam in $input_bam:
+            ln -s '$bam' bam_inputs/catalog.bam &&
+        #end for
+    #else
+        >&2 echo "exactly one (merged) bam file is needed in denovo mode if no population map is given" &&
+        exit 1 &&
+    #end if
+#else
+    @BAM_INPUT@
+#end if
+
+gstacks
+
+#if $mode_cond.mode_select == "denovo":
+    -P bam_inputs
+    $mode_cond.ignore_pe_reads
+    #if $mode_cond.advanced_cond.advanced_select == "yes":
+        --kmer-length $mode_cond.advanced_cond.kmer_length
+        --max-debruijn-reads $mode_cond.advanced_cond.max_debruijn_reads
+        --min-kmer-cov $mode_cond.advanced_cond.min_kmer_cov
+        $mode_cond.advanced_cond.write_alignments
+    #end if
+#else:
+    #if $popmap
+        -I bam_inputs
+    #else
+        $bamlist
+    #end if
+    #if $mode_cond.paired_cond.paired_select == ''
+        $mode_cond.paired_cond.rm_unpaired_reads
+        $mode_cond.paired_cond.rm_pcr_duplicates
+    #else:
+        $mode_cond.paired_cond.paired_select
+    #end if
+    #if $mode_cond.advanced_cond.advanced_select == "yes":
+        --min-mapq $mode_cond.advanced_cond.min_mapq
+        --max-clipped $mode_cond.advanced_cond.max_clipped
+        --max-insert-len $mode_cond.advanced_cond.max_insert_len
+        $mode_cond.advanced_cond.details
+        --phasing-cooccurrences-thr-range $mode_cond.advanced_cond.phasing_cooccurrences_thr_min,$mode_cond.advanced_cond.phasing_cooccurrences_thr_max
+        $mode_cond.advanced_cond.phasing_dont_prune_hets
+    #end if
+#end if
+#if $popmap
+    -M '$popmap'
+#end if
+-O stacks_outputs
+-t \${GALAXY_SLOTS:-1}
+
+##Model options:
+--model $model_cond.model
+--var-alpha $model_cond.var_alpha
+--gt-alpha $model_cond.gt_alpha
+
+
+## the bam files generated by gstacks (--write-alignments) are seemingly buggy
+## (https://groups.google.com/d/msg/stacks-users/CazwJY1DPGA/7vuahiB2GgAJ)
+## so we fix them temporarily by piping them through samtools view (disabling all
+## exit codes and stderr output) this adds the samtools requirement
+## for later versions where this is fixed the output bam files could just be moved
+## to stacks_outputs if this is still necessary
+#if $mode_cond.mode_select == "denovo" and $mode_cond.advanced_cond.advanced_select == "yes" and $mode_cond.advanced_cond.write_alignments != ""
+    #if $popmap:
+        && for b in bam_inputs/*alns.bam; do (samtools view -b "\$b" || true) 2> /dev/null > stacks_outputs/\$(basename "\$b"); done
+    #else
+        && (samtools view -b bam_inputs/alignments.bam || true) 2> /dev/null > stacks_outputs/alignments.bam
+    #end if
+#end if
+
+
+## annoyingly gstacks creates stacks_output/gstacks.log
+## instead of just writing to stderr as the other tools
+## hence we do not use the tokens and return populations.log as log file and take the stderr
+#if $output_log
+    && mv stacks_outputs/gstacks.log $output_log
+#end if
+
+@EXTRACT_VCF@
+
+## TODO extract individual distributions from stacks_outputs/gstacks.log.distribs
+## alternative extra tool
+## for i in \$(stacks-dist-extract stacks_outputs/gstacks.log.distribs)
+## do
+##     stacks-dist-extract stacks_outputs/gstacks.log.distribs $i > stacks_outputs/gstacks.log.\$i.tsv
+## done
+## TODO make optional output collection
+    ]]></command>
+
+    <inputs>
+        <expand macro="bam_input_macro"/>
+        <param name="popmap" type="data" format="tabular,txt" label="Population map" help="If set, matching will be done only for samples listed in this file" optional="true" argument="-M" />
+
+        <conditional name="mode_cond">
+            <param name="mode_select" type="select" label="Mode">
+                <option value="denovo" selected="true">De novo mode</option>
+                <option value="refbased">Reference-based</option>
+            </param>
+            <when value="denovo">
+                <param argument="--ignore-pe-reads" name="ignore_pe_reads" type="boolean" checked="false" truevalue="--ignore-pe-reads" falsevalue="" label="Ignore paired-end reads" help="ignore paired-end reads even if present in the input" />
+                <conditional name="advanced_cond">
+                    <param name="advanced_select" type="select" label="Advanced options">
+                        <option value="no">No</option>
+                        <option value="yes">Yes</option>
+                    </param>
+                    <when value="yes">
+                        <param argument="--kmer-length" name="kmer_length" type="integer" value="31" min="2" max="31" label="K-mer length for the de Bruijn graph" />
+                        <param argument="--max-debruijn-reads" name="max_debruijn_reads" type="integer" value="1000" min="1" label="Maximum number of reads to use in the de Bruijn graph" />
+                        <param argument="--min-kmer-cov" name="min_kmer_cov" type="integer" value="2" label="Minimum coverage to consider a kmer" />
+                        <param argument="--write-alignments" name="write_alignments" type="boolean" checked="false" truevalue="--write-alignments" falsevalue="" label="save read alignments" help="heavy BAM files"/>
+                    </when>
+                    <when value="no"/>
+                </conditional>
+            </when>
+            <when value="refbased">
+                <conditional name="paired_cond">
+                    <param name="paired_select" type="select" label="Paired end options" help="select single/paired for single end data or to select advanced paired end options, --unpaired: treat reverse reads as if they were forward reads; --ignore-pe-reads: ignore paired-end reads even if present in the input">
+                        <option value="" selected="true">single/paired</option>
+                        <option value="--unpaired" selected="true">ignore read pairing (--unpaired)</option>
+                        <option value="--ignore-pe-reads" selected="true">ignore paired-end reads (--ignore-pe-reads)</option>
+                    </param>
+                    <when value="">
+                        <param argument="--rm-unpaired-reads" name="rm_unpaired_reads" type="boolean" checked="false" truevalue="--rm-unpaired-reads" falsevalue="" label="Discard unpaired reads" />
+                        <param argument="--rm-pcr-duplicates" name="rm_pcr_duplicates" type="boolean" checked="false" truevalue="--rm-pcr-duplicates" falsevalue="" label="Remove read pairs of the same sample that have the same insert length" help="implies --rm-unpaired-reads" />
+                    </when>
+                    <when value="--unpaired"/>
+                    <when value="--ignore-pe-reads"/>
+                </conditional>
+                <conditional name="advanced_cond">
+                    <param name="advanced_select" type="select" label="Advanced options">
+                        <option value="no">No</option>
+                        <option value="yes">Yes</option>
+                    </param>
+                    <when value="yes">
+                        <param argument="--min-mapq" name="min_mapq" type="integer" value="10" min="0" max="255" label="Minimum PHRED-scaled mapping quality to consider a read" />
+                        <param argument="--max-clipped" name="max_clipped" type="float" value="0.2" min="0.0" max="1.1" label="Maximum soft-clipping level" help="in fraction of read length" />
+                        <param argument="--max-insert-len" name="max_insert_len" type="integer" value="1000" min="0" label="Maximum allowed sequencing insert length" />
+                        <param argument="--details" type="boolean" checked="false" truevalue="--details" falsevalue="" label="Write a heaview output" />
+                        <param name="phasing_cooccurrences_thr_min" type="integer" value="1" min="0" label="Edge coverage min" help="(--phasing-cooccurrences-thr-range)" />
+                        <param name="phasing_cooccurrences_thr_max" type="integer" value="2" min="0" label="Edge coverage max" help="range of edge coverage thresholds to iterate over when building the graph of allele cooccurrences for SNP phasing (--phasing-cooccurrences-thr-range)"/>
+                        <param argument="--phasing-dont-prune-hets" name="phasing_dont_prune_hets" type="boolean" checked="false" truevalue="--phasing-dont-prune-hets" falsevalue="" label="Don't try to ignore dubious heterozygote genotypes during phasing" />
+                    </when>
+                    <when value="no"/>
+                </conditional>
+            </when>
+        </conditional>
+
+        <conditional name="model_cond">
+            <param argument="--model" type="select" label="Model to use to call variants and genotypes">
+                <option value="marukilow" selected="true">marukilow</option>
+                <option value="marukihigh">marukihigh</option>
+                <option value="snp">snp</option>
+            </param>
+            <when value="marukilow">
+                <expand macro="variant_calling_options_vg" varalpha_default="0.01"/>
+            </when>
+            <when value="marukihigh">
+                <expand macro="variant_calling_options_vg"/>
+            </when>
+            <when value="snp">
+                <expand macro="variant_calling_options_vg"/>
+            </when>
+        </conditional>
+        <param name="add_log_distribs" type="boolean" checked="false" truevalue="yes" falsevalue="no" label="Add log distribs output as dataset" />
+        <expand macro="in_log"/>
+    </inputs>
+    <outputs>
+        <expand macro="out_log"/>
+        <expand macro="gstacks_outputs_full_macro"/>
+    </outputs>
+
+    <tests>
+        <!-- denovomode, w popmap -->
+        <test expect_num_outputs="3">
+            <param name="input_bam" ftype="bam" value="tsv2bam/PopA_01.matches.bam,tsv2bam/PopA_02.matches.bam"/>
+            <param name="popmap" ftype="tabular" value="denovo_map/popmap_cstacks.tsv" />
+            <conditional name="mode_cond">
+                <param name="mode_select" value="denovo"/>
+            </conditional>
+            <param name="add_log" value="yes" />
+            <param name="add_log_distribs" value="yes" />
+            <output name="output_log" ftype="txt" file="gstacks/gstacks.log" lines_diff="8"/>
+            <output name="distribs" ftype="txt" file="gstacks/gstacks.log.distribs" compare="sim_size"/>
+            <output_collection name="gstacks_out" type="list" count="2">
+                <element name="catalog.calls.vcf" file="gstacks/catalog.calls.vcf" ftype="vcf" lines_diff="2"/>
+                <element name="catalog.fa.gz" file="gstacks/catalog.fa.gz" ftype="fasta.gz"/>
+            </output_collection>
+        </test>
+        <!-- denovomode, w popmap, write alignments -->
+        <test expect_num_outputs="3">
+            <param name="input_bam" ftype="bam" value="tsv2bam/PopA_01.matches.bam,tsv2bam/PopA_02.matches.bam"/>
+            <param name="popmap" ftype="tabular" value="denovo_map/popmap_cstacks.tsv" />
+            <conditional name="mode_cond">
+                <param name="mode_select" value="denovo"/>
+                <conditional name="advanced_cond">
+                    <param name="advanced_select" value="yes"/>
+                    <param name="write_alignments" value="--write-alignments" />
+                </conditional>
+            </conditional>
+            <param name="add_log" value="yes" />
+            <assert_command>
+                <has_text text="--write-alignments" />
+            </assert_command>
+            <output name="output_log" ftype="txt"><assert_contents><has_text text="done." /></assert_contents></output>
+            <output_collection name="gstacks_out" type="list" count="2"/>
+            <output_collection name="gstacks_alns_out" type="list" count="2">
+                <element name="PopA_01" file="gstacks/PopA_01.alns.bam" ftype="bam" />
+                <element name="PopA_02" file="gstacks/PopA_02.alns.bam" ftype="bam" />
+            </output_collection>
+        </test>
+        <!-- denovomode, wo popmap (allows for only one input), ignore PE, advanced, alt model -->
+        <test expect_num_outputs="3">
+            <param name="input_bam" value="tsv2bam/PopA_01.matches.bam" ftype="bam"/>
+            <conditional name="mode_cond">
+                <param name="mode_select" value="denovo" />
+                <param name="ignore_pe_reads" value="--ignore-pe-reads" />
+                <conditional name="advanced_cond">
+                    <param name="advanced_select" value="yes"/>
+                    <param name="kmer_length" value="23" />
+                    <param name="max_debruijn_reads" value="666"/>
+                    <param name="min_kmer_cov" value="3" />
+                    <param name="write_alignments" value="--write-alignments" />
+                </conditional>
+            </conditional>
+            <conditional name="model_cond">
+                <param name="model" value="marukihigh"/>
+                <param name="var_alpha" value="0.1" />
+                <param name="gt_alpha" value="0.1" />
+            </conditional>
+            <param name="add_log" value="yes" />
+            <assert_command>
+                <has_text text="--ignore-pe-reads" />
+                <has_text text="--rm-pcr-duplicates" />
+                <has_text text="--kmer-length 23" />
+                <has_text text="--max-debruijn-reads 666" />
+                <has_text text="--min-kmer-cov 3" />
+                <has_text text="--write-alignments" />
+                <has_text text="--model marukihigh" />
+                <has_text text="--var-alpha 0.1" />
+                <has_text text="--gt-alpha 0.1" />
+            </assert_command>
+            <output name="output_log" ftype="txt"><assert_contents><has_text text="done." /></assert_contents></output>
+            <output_collection name="gstacks_out" type="list" count="2"/>
+            <output name="gstacks_aln_out" ftype="bam" file="gstacks/alignments.bam" />
+        </test>
+        <!-- refbased wo popmap, paired options, removing all unpaired reads results in an error -->
+        <test expect_failure="true" expect_exit_code="1">
+            <param name="input_bam" value="tsv2bam/PopA_01.bam,tsv2bam/PopA_02.bam"/>
+            <conditional name="mode_cond">
+                <param name="mode_select" value="refbased"/>
+                <conditional name="paired_cond">
+                    <param name="paired_select" value=""/>
+                    <!--<param name="rm_unpaired_reads" value="\-\-rm-unpaired-reads" /> removes to much of the test data and gstacks fails-->
+                    <param name="rm_pcr_duplicates" value="--rm-pcr-duplicates" />
+                </conditional>
+            </conditional>
+            <param name="add_log" value="yes" />
+            <assert_command>
+                <has_text text="-I bam_inputs" />
+                <not_has_text text="-B " />
+                <has_text text="--rm-unpaired-reads" />
+                <has_text text="--rm-pcr-duplicates" />
+            </assert_command>
+        </test>
+        <!-- refbased w popmap (here bam names need to be equal to sample names in popmap), \-\-unpaired, advanced, snp model -->
+        <test expect_num_outputs="2">
+            <param name="input_bam" ftype="bam" value="tsv2bam/PopA_01.bam,tsv2bam/PopA_02.bam"/>
+            <param name="popmap" ftype="tabular" value="denovo_map/popmap_cstacks.tsv" />
+            <conditional name="mode_cond">
+                <param name="mode_select" value="refbased"/>
+                <conditional name="paired_cond">
+                    <param name="paired_select" value="--unpaired"/>
+                </conditional>
+                <conditional name="advanced_cond">
+                    <param name="advanced_select" value="yes" />
+                    <param name="min_mapq" value="23" />
+                    <param name="max_clipped" value="0.23" />
+                    <param name="max_insert_len" value="666" />
+                    <param name="details" value="--detailed"/>
+                    <param name="phasing_cooccurrences_thr_min" value="2"/>
+                    <param name="phasing_cooccurrences_thr_max" value="3"/>
+                    <param name="phasing_dont_prune_hets" value="--phasing-dont-prune-hets" />
+                </conditional>
+            </conditional>
+            <param name="model_cond|model" value="snp"/>
+            <param name="model_cond|gt_alpha" value="0.1" />
+            <param name="model_cond|var_alpha" value="0.1" />
+            <param name="add_log" value="yes" />
+            <assert_command>
+                <not_has_text text="-I bam_inputs" />
+                <has_text text="-B " />
+                <has_text text="--unpaired" />
+                <has_text text="--min-mapq 23" />
+                <has_text text="--max-clipped 0.23" />
+                <has_text text="--max-insert-len 666" />
+                <has_text text="--detailed" />
+                <has_text text="--phasing-cooccurrences-thr-range 2,3" />
+                <has_text text="--phasing-dont-prune-hets" />
+                <has_text text="--model snp" />
+                <has_text text="--gt-alpha 0.1" />
+            </assert_command>
+            <output name="output_log" ftype="txt"><assert_contents><has_text text="done." /></assert_contents></output>
+            <output_collection name="gstacks_out" type="list" count="2"/>
+        </test>
+        <!-- refbased wo popmap (here bam names don't matter), \-\-ignorepe -->
+        <test expect_num_outputs="2">
+            <param name="input_bam" ftype="bam" value="tsv2bam/PopA_01.bam,tsv2bam/PopA_02.bam"/>
+            <conditional name="mode_cond">
+                <param name="mode_select" value="refbased"/>
+                <conditional name="paired_cond">
+                    <param name="paired_select" value="--ignore-pe-reads"/>
+                </conditional>
+            </conditional>
+            <param name="add_log" value="yes" />
+            <assert_command>
+                <has_text text="-I bam_inputs" />
+                <not_has_text text="-B " />
+                <has_text text="--ignore-pe-reads" />
+            </assert_command>
+            <output name="output_log"><assert_contents><has_text text="gstacks is done." /></assert_contents></output>
+            <output_collection name="gstacks_out" type="list" count="2"/>
+        </test>
+    </tests>
+
+    <help>
+<![CDATA[
+.. class:: infomark
+
+**What it does**
+
+For de novo analyses, this program will pull in paired-end reads, if available,
+assemble the paired-end contig and merge it with the single-end locus, align
+reads to the locus, and call SNPs.
+
+For reference-aligned analyses, this program will build loci from the single
+and/or paired-end reads before calling SNPs. The single- and paired-end reads
+must be aligned and stored together in the intput BAM or SAM files and the
+reads must be sorted. The gstacks program will detect if single- or paired-end
+reads are present.
+
+In either mode, gstacks is able to remove PCR duplicates if requested.
+
+--------
+
+**Input files**
+
+If a population map is given BAM records must be assigned to samples using BAM "reads groups"
+(gstacks uses the ID/identifier and SM/sample name fields). Read groups
+must be consistent if repeated different files.
+Otherwise read groups are unneeded and ignored.
+
+**Output files**
+
+- Assembled contigs and variant sites
+
+- Optional outputs: Read alignments and log.distribs
+
+@STACKS_INFOS@
+]]>
+    </help>
+    <expand macro="citation" />
+</tool>