--- a/concoct.xml	Fri Jul 01 14:14:44 2022 +0000
+++ b/concoct.xml	Thu Jul 07 08:33:35 2022 +0000
@@ -1,99 +1,140 @@
 <tool id="concoct" name="CONCOCT" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@PROFILE@">
-    <description>metagenome binning</description>
+    <description>for metagenome binning</description>
     <macros>
         <import>macros.xml</import>
     </macros>
     <expand macro="requirements"/>
     <command detect_errors="exit_code"><![CDATA[
-#set pca_components_file_name = 'PCA_components_data_gt' + str($advanced.length_threshold) + '.csv'
-#set pca_transformed_file_name = 'PCA_transformed_data_gt' + str($advanced.length_threshold) + '.csv'
-#set clustering_file_name = 'clustering_gt' + str($advanced.length_threshold) + '.csv'
-
 ## CONCOCT doesn't handle gzipped files.
 #if $composition_file.ext.endswith(".gz")
-    gunzip -c '$composition_file' > composition_file.fa &&
+    gunzip -c '$composition_file' > 'composition_file.fa' &&
 #else:
-    ln -s '$composition_file' composition_file.fa &&
+    ln -s '$composition_file' 'composition_file.fa' &&
 #end if
 
 mkdir outdir &&
 concoct
---coverage_file '$coverage_file'
---composition_file composition_file.fa
---clusters $advanced.clusters
---kmer_length $advanced.kmer_length
---threads \${GALAXY_SLOTS:-4}
---length_threshold $advanced.length_threshold
---read_length $advanced.read_length
---total_percentage_pca $advanced.total_percentage_pca
---basename 'outdir/'
---seed $advanced.seed
---iterations $advanced.iterations
---epsilon $advanced.epsilon
-$advanced.no_cov_normalization
-$advanced.no_total_coverage
---no_original_data
-$advanced.converge_out
-
-## Convert all CONCOCT .csv outputs to tabular.
-&& sed 's/\("\([^"]*\)"\)\?,/\2\t/g' outdir/$pca_components_file_name > '$output_pca_components'
-&& sed 's/\("\([^"]*\)"\)\?,/\2\t/g' outdir/$pca_transformed_file_name > '$output_pca_transformed'
-&& sed 's/\("\([^"]*\)"\)\?,/\2\t/g' outdir/$clustering_file_name > '$output_clustering'
-#if str($advanced.output_process_log) == 'yes':
-    && mv outdir/log.txt '$process_log'
-#end if
+    --coverage_file '$coverage_file'
+    --composition_file 'composition_file.fa'
+    --clusters $advanced.clusters
+    --kmer_length $advanced.kmer_length
+    --threads \${GALAXY_SLOTS:-4}
+    --length_threshold $advanced.length_threshold
+    --read_length $advanced.read_length
+    --total_percentage_pca $advanced.total_percentage_pca
+    --basename 'outdir/'
+    --seed $advanced.seed
+    --iterations $advanced.iterations
+    $advanced.no_cov_normalization
+    $output.no_total_coverage
+    --no_original_data
+    $output.converge_out
     ]]></command>
     <inputs>
-        <param argument="--coverage_file" type="data" format="tabular" label="Tabular coverage file" help="Columns correspond to samples and rows to contigs"/>
-        <param argument="--composition_file" type="data" format="fasta,fasta.gz" label="Fasta file" help="Used to calculate the kmer composition (the genomic signature) of each contig"/>
+        <param argument="--coverage_file" type="data" format="tabular" label="Coverage file" help="Table where each row correspond to a contig, and each column correspond to a sample. The values are the average coverage for this contig in that sample"/>
+        <param argument="--composition_file" type="data" format="fasta,fasta.gz" label="Composition file with sequences" help="It is named the composition file since it is used to calculate the kmer composition (the genomic signature) of each contig."/>
         <section name="advanced" title="Advanced options">
-            <param argument="--clusters" type="integer" value="400" label="Maximum number of clusters for the Variational Gaussian Mixture Model algorithm"/>
-            <param argument="--kmer_length" type="integer" value="4" label="Kmer length"/>
-            <param argument="--length_threshold" type="integer" value="1000" label="Sequence length threshold" help="Contigs shorter than this value will not be included"/>
-            <param argument="--read_length" type="integer" value="100" label="Read length for coverage"/>
-            <param argument="--total_percentage_pca" type="integer" value="100" label="Percentage of variance explained by the principal components for the combined data"/>
-            <param argument="--seed" type="integer" min="0" value="1" label="Integer to use as seed for clustering" help="Zero value will use random seed"/>
-            <param argument="--iterations" type="integer" value="500" label="Maximum number of iterations for the Variational Bayes Gaussian Mixture Models"/>
-            <param argument="--epsilon" type="float" value="0.000001" label="Epsilon for the Variational Gaussian Mixture Model algorithm"/>
-            <param argument="--no_cov_normalization" type="boolean" truevalue="--no_cov_normalization" falsevalue="" checked="false" label="Skip normalization and only do log transorm of the coverage?" help="By default, the coverage is normalized for samples, then normalized for contigs and finally log transformed"/>
-            <param argument="--no_total_coverage" type="boolean" truevalue="--no_total_coverage" falsevalue="" checked="false" label="Eliminate the total coverage column from the coverage data matrix?" help="By default, total coverage is included, independently of coverage normalization but previous to log transformation"/>
-            <param argument="--converge_out" type="boolean" truevalue="--converge_out" falsevalue="" checked="false" label="Output convergence information?"/>
-            <param name="output_process_log" type="select" label="Output process log file?">
-                <option value="no" selected="true">No</option>
-                <option value="yes">Yes</option>
-            </param>
+            <param argument="--clusters" type="integer" min="0" value="400" label="Maximum number of clusters for the Variational Gaussian Mixture Model (VGMM) algorithm"/>
+            <param argument="--kmer_length" type="integer" min="0" value="4" label="Kmer length"/>
+            <param argument="--length_threshold" type="integer" min="0" value="1000" label="Sequence length threshold" help="Contigs shorter than this value will not be included"/>
+            <param argument="--read_length" type="integer" min="0" value="100" label="Read length for coverage"/>
+            <param argument="--total_percentage_pca" type="integer" min="0" value="100" label="Percentage of variance explained by the principal components for the combined data"/>
+            <param argument="--seed" type="integer" min="0" value="1" label="Seed for clustering" help="Zero value will use random seed"/>
+            <param argument="--iterations" type="integer" min="0" value="500" label="Maximum number of iterations for the Variational Bayes Gaussian Mixture Models (VBGMM)"/>
+            <param argument="--no_cov_normalization" type="boolean" truevalue="--no_cov_normalization" falsevalue="" checked="false" label="Skip normalization and only do log transorm of the coverage?" help="By default, the coverage is normalized for samples, then normalized for contigs and finally log transformed. By setting this flag you skip the normalization and only do log transorm of the   coverage."/>
+        </section>
+        <section name="output" title="Output">
+            <param argument="--no_total_coverage" type="boolean" truevalue="--no_total_coverage" falsevalue="" checked="false" label="Eliminate the total coverage column from the coverage data matrix?" help="By default, total coverage is included, independently of coverage normalization but previous to log transformation. Use this tag to escape this behaviour."/>
+            <param argument="--converge_out" type="boolean" truevalue="--converge_out" falsevalue="" checked="false" label="Write convergence information to files?"/>
+            <param name="log" type="boolean" checked="false" label="Output process log file?"/>
         </section>
     </inputs>
     <outputs>
-        <data name="process_log" format="txt" label="${tool.name} on ${on_string} (process log)">
-            <filter>advanced['output_process_log'] == 'yes'</filter>
+        <data name="output_clustering" format="csv" from_work_dir="outdir/clustering_gt*" label="${tool.name} on ${on_string}: Clusters"/>
+        <data name="process_log" format="txt" from_work_dir="outdir/log.txt" label="${tool.name} on ${on_string}: Log">
+            <filter>output['log']</filter>
         </data>
-        <data name="output_pca_components" format="tabular" label="${tool.name} on ${on_string} (PCA components)"/>
-        <data name="output_pca_transformed" format="tabular" label="${tool.name} on ${on_string} (PCA transformed)"/>
-        <data name="output_clustering" format="tabular" label="${tool.name} on ${on_string} (Clusters)"/>
+        <data name="output_pca_components" format="csv" from_work_dir="outdir/PCA_components_data_gt*" label="${tool.name} on ${on_string}: PCA components"/>
+        <data name="output_pca_transformed" format="csv" from_work_dir="outdir/PCA_transformed_data_gt*" label="${tool.name} on ${on_string}: PCA transformed clusters"/>
     </outputs>
     <tests>
         <test expect_num_outputs="4">
-            <param name="coverage_file" value="input1.tabular" ftype="tabular"/>
-            <param name="composition_file" value="input1.fa.gz" ftype="fasta.gz"/>
-            <param name="output_process_log" value="yes"/>
-            <output name="process_log" file="process_log.txt" ftype="txt" compare="re_match"/>
-            <output name="output_pca_components" ftype="tabular">
+            <param name="coverage_file" value="coverage" ftype="tabular"/>
+            <param name="composition_file" value="composition.fa" ftype="fasta"/>
+            <section name="advanced">
+                <param name="clusters" value="400"/>
+                <param name="kmer_length" value="4"/>
+                <param name="length_threshold" value="1000"/>
+                <param name="read_length" value="100"/>
+                <param name="total_percentage_pca" value="100"/>
+                <param name="seed" value="1"/>
+                <param name="iterations" value="500"/>
+                <param name="no_cov_normalization" value=""/>
+            </section>
+            <section name="output">
+                <param name="no_total_coverage" value=""/>
+                <param name="converge_out" value=""/>
+                <param name="log" value="true"/>
+            </section>
+            <output name="process_log" ftype="txt" compare="contains">
                 <assert_contents>
-                    <has_size value="367636"/>
-                    <has_text text="7377051e-02"/>
+                    <has_size value="786"/>
+                    <has_text text="CONCOCT Finished"/>
+                </assert_contents>
+            </output>
+            <output name="output_pca_components" ftype="csv">
+                <assert_contents>
+                    <has_size value="362924" delta="10"/>
+                    <has_text text="-5.90697200e-02"/>
+                </assert_contents>
+            </output>
+            <output name="output_pca_transformed" ftype="csv">
+                <assert_contents>
+                    <has_size value="834200" delta="10"/>
+                    <has_text text="contig-21000001"/>
                 </assert_contents>
             </output>
-            <output name="output_pca_transformed" ftype="tabular">
+            <output name="output_clustering" ftype="csv">
                 <assert_contents>
-                    <has_size value="737926"/>
-                    <has_text text="NODE_103_length_20202_cov_8.395357.0"/>
+                    <has_size value="6923" delta="10"/>
+                    <has_text text="contig-21000001,"/>
                 </assert_contents>
             </output>
-            <output name="output_clustering" ftype="tabular">
+        </test>
+        <test expect_num_outputs="3">
+            <param name="coverage_file" value="coverage" ftype="tabular"/>
+            <param name="composition_file" value="composition.fa.gz" ftype="fasta.gz"/>
+            <section name="advanced">
+                <param name="clusters" value="400"/>
+                <param name="kmer_length" value="4"/>
+                <param name="length_threshold" value="1000"/>
+                <param name="read_length" value="100"/>
+                <param name="total_percentage_pca" value="100"/>
+                <param name="seed" value="1"/>
+                <param name="iterations" value="500"/>
+                <param name="no_cov_normalization" value=""/>
+            </section>
+            <section name="output">
+                <param name="no_total_coverage" value=""/>
+                <param name="converge_out" value=""/>
+                <param name="log" value="false"/>
+            </section>
+            <output name="output_pca_components" ftype="csv">
                 <assert_contents>
-                    <has_size value="12167"/>
-                    <has_text text="NODE_103_length_20202_cov_8.395357"/>
+                    <has_size value="362924" delta="10"/>
+                    <has_text text="-5.90697200e-02"/>
+                </assert_contents>
+            </output>
+            <output name="output_pca_transformed" ftype="csv">
+                <assert_contents>
+                    <has_size value="834200" delta="10"/>
+                    <has_text text="contig-21000001"/>
+                </assert_contents>
+            </output>
+            <output name="output_clustering" ftype="csv">
+                <assert_contents>
+                    <has_size value="6923" delta="10"/>
+                    <has_text text="contig-21000001,"/>
                 </assert_contents>
             </output>
         </test>
@@ -103,7 +144,7 @@
 
 CONCOCT (Clustering cONtigs with COverage and ComposiTion) performs unsupervised binning of metagenomic contigs by
 using nucleotide composition - kmer frequencies - and coverage data for multiple samples.  CONCOCT can accurately
-(up to species level) bin metagenomic contigs. 
+(up to species level) bin metagenomic contigs.
 
 The tool accepts 2 inputs; a tabular file where each row corresponds to a contig and each column corresponds to a
 sample (the values are the average coverage for this contig in that sample) and a file containing sequences in