Ampvis subset allow file input for taxa selection (#6463)

* input update and test case * fixed test * bump --------- Co-authored-by: Björn Grüning <bjoern@gruenings.eu>
galaxyproject · Oct 25, 2024 · d929359 · d929359
1 parent 938f808
commit d929359
Show file tree

Hide file tree

Showing 2 changed files with 72 additions and 10 deletions.
diff --git a/tools/ampvis2/subset_taxa.xml b/tools/ampvis2/subset_taxa.xml
@@ -1,4 +1,4 @@
-<tool id="ampvis2_subset_taxa" name="ampvis2 subset data" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@PROFILE@" license="MIT">
+<tool id="ampvis2_subset_taxa" name="ampvis2 subset data" version="@TOOL_VERSION@+galaxy1" profile="@PROFILE@" license="MIT">
     <description>by taxonomy</description>
     <macros>
         <import>macros.xml</import>
@@ -11,33 +11,62 @@
         <configfile name="rscript"><![CDATA[
             library(ampvis2, quietly = TRUE)
             data <- readRDS("$data")
-            #set tax_vector = 'c("' + '", "'.join(str($tax_vector).split(",")) + '")'
-            data <- amp_subset_taxa(
+
+            #if $select_param == "option_input_selected_file"
+                ## Define the file path (assuming it's a text file with one value per line)
+                file_path <- "$selected_taxonomy_list"
+
+                ## Read the file into a character vector, where each line represents a value
+                lines <- readLines(file_path)
+
+                ## Remove any leading or trailing whitespace (if necessary)
+                tax_vector <- trimws(lines)
+            #else
+                tax_vector <- unlist(strsplit("$tax_vector", ","))
+            #end if
+                data <- amp_subset_taxa(
                 data,
-                tax_vector = $tax_vector,
+                tax_vector = tax_vector,
                 normalise = $normalise,
                 remove = $remove
-            )
+                )     
+
             saveRDS(data, "$ampvis")
             @SAVE_TAX_LIST@
         ]]></configfile>
     </configfiles>
     <inputs>
         <expand macro="rds_input_macro"/>
         <!-- add validator -->
-        <param name="taxonomy_list" type="data" optional="false" format="tabular" label="Taxonomy list" help="Generated with ampvis2: load"/>
-        <expand macro="taxonomy_select_macro" argument="tax_vector" label="Taxa to keep"/>
+        <conditional name="taxonomy_select_type">
+            <param name="select_param" type="select" label="Choose taxa input option">
+                <option value="option_input_file">Select taxa from taxonomy list file</option>
+                <option value="option_input_selected_file">Provide a file with selected taxa</option>
+            </param>
+
+            <when value="option_input_selected_file">
+                <param name="selected_taxonomy_list" type="data" optional="false" format="tabular" label="Selected taxonomy list" help="A file containing the taxa to subsample. One taxa per line."/>
+            </when>
+
+            <when value="option_input_file">
+                <param name="taxonomy_list" type="data" optional="false" format="tabular" label="Taxonomy list" help="Generated with ampvis2: load"/>
+                <expand macro="taxonomy_select_macro" argument="tax_vector" label="Taxa to keep"/>
+            </when>
+        </conditional>
+
         <expand macro="normalise_macro"/>
         <param argument="remove" type="boolean" truevalue="TRUE" falsevalue="FALSE" checked="false" label="Remove" help="Selected taxa will be removed instead of being the only ones kept in the data."/>
     </inputs>
     <outputs>
         <data name="ampvis" format="ampvis2"/>
         <data name="taxonomy_list_out" format="tabular" label="${tool.name} on ${on_string}: taxonomy list"/>
     </outputs>
+    <!-- Test cases -->
     <tests>
-        <!-- defaults -->
+        <!-- Test case for using comma-separated taxonomy list -->
         <test expect_num_outputs="2">
             <param name="data" value="AalborgWWTPs.rds" ftype="ampvis2"/> 
+            <param name="select_param" value="option_input_file"/>
             <param name="taxonomy_list" value="AalborgWWTPs-taxonomy.list"/> 
             <param name="tax_vector" value="f__Caldilineaceae,p__Elusimicrobia"/>
             <output name="ampvis" value="AalborgWWTPs-subset_taxa.rds" ftype="ampvis2" compare="sim_size"/>
@@ -48,9 +77,11 @@
                 </assert_contents>
             </output>
         </test>
-        <!-- defaults -->
+
+        <!-- Test case for using comma-separated taxonomy list and remove=true -->
         <test expect_num_outputs="2">
             <param name="data" value="AalborgWWTPs.rds" ftype="ampvis2"/> 
+            <param name="select_param" value="option_input_file"/>
             <param name="taxonomy_list" value="AalborgWWTPs-taxonomy.list"/> 
             <param name="tax_vector" value="f__Caldilineaceae,p__Elusimicrobia"/>
             <param name="remove" value="true"/>
@@ -62,6 +93,21 @@
                 </assert_contents>
             </output>
         </test>
+
+        <!-- Test case for using selected taxonomy list from a file -->
+        <test expect_num_outputs="2">
+            <param name="data" value="AalborgWWTPs.rds" ftype="ampvis2"/> 
+            <param name="select_param" value="option_input_selected_file"/>
+            <param name="selected_taxonomy_list" value="AalborgWWTPs-selected.taxonomy.list"/> 
+            <param name="remove" value="true"/>
+            <output name="ampvis" value="AalborgWWTPs-subset_taxa.rds" ftype="ampvis2" compare="sim_size"/>
+            <output name="taxonomy_list_out" ftype="tabular">
+                <assert_contents>
+                    <not_has_text text="f__Caldilineaceae"/>
+                    <not_has_text text="p__Elusimicrobia"/>
+                </assert_contents>
+            </output>
+        </test>
     </tests>
     <help><![CDATA[
 What it does
@@ -73,12 +119,26 @@ The Galaxy tool calls the `amp_subset_taxa
 <https://kasperskytte.github.io/ampvis2/reference/amp_filter_taxa.html>`_ function
 of the ampvis2 package.
 
+This tool provides two options for specifying the `subset`:
+
+1. **Provide taxa names matching the taxonomy table**:
+
 The taxonomy subset is done by providing a tax_vector of taxa names which are
 then matched to the taxonomy table, where all other taxa not matching the
 "Taxa to keep" (``tax_vector``) are removed. If "Remove" is enabled, then the
 matching taxa are the ones being removed instead. The taxa names 
 will be matched in all columns of the taxonomy table.
 
+2. **Provide a file of selected taxonomies**:
+
+In this option, you can upload a file containing the taxa you wish to subset. 
+The file should contain one taxon name per line. 
+This option is useful when you have a pre-defined list of taxa stored in a file.
+This is useful if you want to use the tool in a workflow or if you automatically generate
+a taxonomy selection e.g. with differential abundance tools like MaAsLi2 and you only want to 
+plot these taxa.
+
+
 @HELP_RELATIVE_ABUNDANCES@
 
 Input
@@ -95,4 +155,4 @@ An ampvis2_rds dataset containing subsetted data and an updated taxonomy list da
 
     ]]></help>
     <expand macro="citations"/>
-</tool>
+</tool>
diff --git a/tools/ampvis2/test-data/AalborgWWTPs-selected.taxonomy.list b/tools/ampvis2/test-data/AalborgWWTPs-selected.taxonomy.list
@@ -0,0 +1,2 @@
+f__Caldilineaceae
+p__Elusimicrobia