changeset 1:eee2da03d1f5 draft

"planemo upload for repository https://github.com/COMBAT-TB/tb_variant_filter commit ef1b4164979078a6624c5b81fa156755258fb5f8"
author iuc
date Sat, 22 May 2021 10:13:18 +0000
parents 3b1e7c170b10
children e5c84ba20d99
files tb_variant_filter.xml
diffstat 1 files changed, 23 insertions(+), 6 deletions(-) [+]
line wrap: on
line diff
--- a/tb_variant_filter.xml	Thu Oct 10 17:59:02 2019 -0400
+++ b/tb_variant_filter.xml	Sat May 22 10:13:18 2021 +0000
@@ -1,7 +1,7 @@
-<tool id="tb_variant_filter" name="TB Variant Filter" version="@TOOL_VERSION@+galaxy0" profile="16.04">
+<tool id="tb_variant_filter" name="TB Variant Filter" version="@TOOL_VERSION@+galaxy0" profile="21.01">
     <description>M. tuberculosis H37Rv VCF filter</description>
     <macros>
-        <token name="@TOOL_VERSION@">0.1.3</token>
+        <token name="@TOOL_VERSION@">0.3.0</token>
     </macros>
     <requirements>
         <requirement type="package" version="@TOOL_VERSION@">tb_variant_filter</requirement>
@@ -57,6 +57,7 @@
             <when value="yes">
                 <param argument="--region_filter" type="select" multiple="true" label="Region filters to enable">
                     <!-- if these are changed the code above needs to change to keep the defaults in line with those that are default here -->
+                    <option value="farhat_rlc">Refined Low Confidence regions from Farhat lab</option>
                     <option value="pe_ppe" selected="true">PE/PPE</option>
                     <option value="tbprofiler">TBProfiler antibiotic resistant genes</option>
                     <option value="mtbseq">MTBseq antibiotic resistant genes</option>
@@ -125,6 +126,17 @@
             <param name="filters" value="snv_only_filter" />
             <output name="output1" file="test_output6.vcf" ftype="vcf" />
         </test>
+        <test>
+            <param name="input1" value="test_input1.vcf" ftype="vcf" />
+            <param name="filters" value="region_filter" />
+            <conditional name="filter_options">
+                <param name="show_filter_options" value="yes" />
+                <param name="region_filter" value="farhat_rlc" />
+            </conditional>
+            <assert_stderr>
+                <has_text text="masked: 2" />
+            </assert_stdout>
+        </test>
     </tests>
     <help><![CDATA[
 This tool offers multiple options for filtering variants (in
@@ -133,15 +145,20 @@
 It currently has 5 main modes:
 
 1. Filter by region. Mask out variants in certain regions. Region lists available as:
-    1.  PE/PPE genes from `Fishbein et al 2015 <https://onlinelibrary.wiley.com/doi/full/10.1111/mmi.12981>`_
-    2. `TBProfiler <http://tbdr.lshtm.ac.uk/>`_ list of antibiotic resistant genes
-    3. `MTBseq <https://github.com/ngs-fzb/MTBseq_source>`_ list of antibiotic resistant genes
-    4. `UVP <https://github.com/CPTR-ReSeqTB/UVP>`_ list of repetitive loci in M. tuberculosis genome
+    1.  Refined Low Confidence (RLC) regions from `Marin et al 2021 <https://www.biorxiv.org/content/10.1101/2021.04.08.438862v1.full>`_
+    2.  PE/PPE genes from `Fishbein et al 2015 <https://onlinelibrary.wiley.com/doi/full/10.1111/mmi.12981>`_
+    3. `TBProfiler <http://tbdr.lshtm.ac.uk/>`_ list of antibiotic resistant genes
+    4. `MTBseq <https://github.com/ngs-fzb/MTBseq_source>`_ list of antibiotic resistant genes
+    5. `UVP <https://github.com/CPTR-ReSeqTB/UVP>`_ list of repetitive loci in M. tuberculosis genome
 2. Filter by window around indels. Masks out variants within a certain distance (by default 5 bases) of an insertion or deletion site.
 3. Filter by percentage of alternate allele bases. Mask out variants with less than a minimum percentage (by default 90%) alternative alleles.
 4. Filter by depth of aligned reads.
 5. Filter out all variants that are not SNV (single nucleotide variants).
 
+For region filtering, the default choice is to use the PE/PPE and UVP regions to mask out variants. `Marin et al 2021 <https://www.biorxiv.org/content/10.1101/2021.04.08.438862v1.full>`_
+from Prof Maha Farhat's lab make a persuasive argument that their smaller list of Refined Low Confidence (RLC) regions is a better argument but this work has not yet been peer
+review so it is included as an option that is not currently the default.
+
 When used together the effects of the filters are added (i.e. a variant is masked out if it is masked by any of the filters).
     ]]></help>
     <citations>