diff cfsan_snp_pipeline_filter_snp_regions.xml @ 0:be8994ce4fb6 draft

Uploaded

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/cfsan_snp_pipeline_filter_snp_regions.xml	Thu Oct 12 19:55:24 2023 +0000
@@ -0,0 +1,57 @@
+<tool id="cfsan_snp_pipeline_filter_snp_regions" name="CFSAN SNP Pipeline: filter SNP Regions" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@PROFILE@">
+    <description>based on location</description>
+    <macros>
+        <import>macros.xml</import>
+    </macros>
+    <expand macro="requirements"/>
+    <command detect_errors="exit_code"><![CDATA[
+#if str($reference_cond.reference_source) == 'cached'
+    #set reference = $reference_cond.reference.fields.path
+#else:
+    #set reference = $reference_cond.reference
+#end if
+mkdir -p ./samples/sample/ &&
+ln -s '$calls' ./samples/sample/var.flt.vcf &&
+echo "\$PWD/samples/sample" > ./alignment.txt &&
+cfsan_snp_pipeline filter_regions ./alignment.txt '$reference' &&
+cp ./samples/sample/var.flt_preserved.vcf '$filcalls' &&
+cp ./samples/sample/var.flt_removed.vcf '$excludes'
+    ]]></command>
+    <inputs>
+        <param name="calls" type="data" format="vcf" label="Unfiltered SNP calls from your history"/>
+        <expand macro="reference_cond"/>
+    </inputs>
+    <outputs>
+        <data name="filcalls" format="vcf" label="Filtered SNPs"/>
+        <data name="excludes" format="vcf" label="SNPs excluded by filtering"/>
+    </outputs>
+    <tests>
+        <test>
+            <param name="calls" value="input.vcf" />
+            <param name="reference_source" value="history"/>
+            <param name="reference" value="lambda_virus.fasta"/>
+            <output name="filcalls" value="output_preserved.vcf" lines_diff="6"/>
+            <output name="excludes" value="output_removed.vcf" lines_diff="6"/>
+        </test>
+    </tests>
+    <help><![CDATA[
+**What it does**
+
+Remove abnormally dense SNPs from the input VCF file, save the reserved SNPs
+into a new VCF file, and save the removed SNPs into another VCF file.
+
+**Optional parameters**
+
+ * **force** - Force processing even when result files already exist and are newer than inputs (default: False)
+ * **vcfname** - File name of the input VCF files which must exist in each of the sample directories (default: var.flt.vcf)
+ * **edge_length** - The length of the edge regions in a contig, in which all SNPs will be removed. (default: 500)
+ * **window_size** - The length of the window in which the number of SNPs should be no more than max_num_snp. (default: [1000])
+ * **max_snp** - The maximum number of SNPs allowed in a window.  (default: [3])
+ * **out_group** - Relative or absolute path to the file indicating outgroup samples, one sample ID per line. (default: None)
+
+**More information**
+
+[CFSAN SNP Pipeline documentation](http://snp-pipeline.readthedocs.io/en/latest/index.html">http://snp-pipeline.readthedocs.io/en/latest/index.html)
+    ]]></help>
+    <expand macro="citations"/>
+</tool>