diff windowmasker_mkcounts.xml @ 0:d90ae4a02efd draft default tip

planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/main/tools/windowmasker/ commit d529dafd5a2d3a017de226cd59e42cf630348e06
author iuc
date Thu, 14 Dec 2023 16:04:27 +0000
parents
children
line wrap: on
line diff
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/windowmasker_mkcounts.xml	Thu Dec 14 16:04:27 2023 +0000
@@ -0,0 +1,101 @@
+<tool id="windowmasker_mkcounts" name="WindowMasker mkcounts" version="1.0" profile="22.01">
+    <description>Construct WindowMasker unit counts table</description>
+    <macros>
+        <import>macros.xml</import>
+    </macros>
+    <expand macro="requirements" />
+    <command detect_errors="exit_code">
+<![CDATA[
+    windowmasker -mk_counts -infmt fasta
+        #if $adv.genome_size
+            -genome_size $adv.genome_size
+        #end if
+        #if $adv.t_low
+            -t_low $adv.t_low
+        #end if
+        #if $adv.t_high
+            -t_high $adv.t_high
+        #end if
+        #if $adv.unit
+            -unit $adv.unit
+        #end if
+
+        -sformat ${adv.sformat}
+        ${adv.checkdup}
+        -in '${fasta_input}'
+        -out '${mkcount_output}'
+]]>
+    </command>
+    <inputs>
+        <param argument="-in" name="fasta_input" type="data" format="fasta"
+                label="FASTA sequence file" />
+
+        <section name="adv" title="Advanced options" expanded="false" >
+            <param argument="-checkdup" type="boolean" checked="false"
+                    truevalue="-checkdup" falsevalue=""
+                    label="Check input file for potentially duplicated sequences" />
+
+            <param argument="-genome_size" type="integer" label="Genome size"
+                    min="0" optional="true"
+                    help="Use this genome size to compute the unit length" />
+
+            <param argument="-sformat" type="select" label="Unit counts format"
+                help="The format the unit counts data should be generated" >
+                <option value="ascii">Plain text (ascii)</option>
+                <option value="oascii" selected="true">Optimized text (oascii)</option>
+                <!-- Omit support for non-portable binary formats (binary, obinary) -->
+            </param>
+
+            <param argument="-t_low" type="integer" label="T_low"
+                    min="0" optional="true"
+                    help="Save only units that appears at least this many times" />
+
+            <param argument="-t_high" type="integer" label="T_high"
+                    min="0" optional="true"
+                    help="Set the maximum count value for units" />
+
+            <param argument="-unit" type="integer" label="Unit length"
+                    min="1" max="16" optional="true"
+                    help="Unit length used to mask the genome" />
+        </section>
+    </inputs>
+    <outputs>
+        <data name="mkcount_output" format="txt" />
+    </outputs>
+    <tests>
+        <test>
+            <!-- Test mkcounts with default settings -->
+            <param name="fasta_input" value="chr4_part.fa" ftype="fasta" />
+            <output name="mkcount_output" file="chr4_part.oascii.counts" />
+        </test>
+        <test>
+            <!-- Test mkcounts with custom genome_size -->
+            <param name="fasta_input" value="chr4_part.fa" ftype="fasta" />
+            <param name="genome_size" value="500000" />
+            <output name="mkcount_output" file="chr4_part.500k_size.counts" />
+        </test>
+        <test>
+            <!-- Test mkcounts with ascii output format -->
+            <param name="fasta_input" value="chr4_part.fa" ftype="fasta" />
+            <param name="sformat" value="ascii" />
+            <output name="mkcount_output" file="chr4_part.ascii.counts" />
+        </test>
+        <test>
+            <!-- Test mkcounts with custom unit settings -->
+            <param name="fasta_input" value="chr4_part.fa" ftype="fasta" />
+            <param name="t_low" value="3" />
+            <param name="t_high" value="20" />
+            <param name="unit" value="9" />
+            <output name="mkcount_output" file="chr4_part.custom_unit.counts" />
+        </test>
+    </tests>
+    <help>
+<![CDATA[
+**What it does**
+
+This tool runs `stage 1 <https://www.ncbi.nlm.nih.gov/IEB/ToolBox/CPP_DOC/lxr/source/src/app/winmasker/>`_
+of the WindowMasker analysis to produce a unit counts file for a genome assembly.
+
+    ]]></help>
+    <expand macro="citations" />
+</tool>