--- a/hicBuildMatrix.xml	Tue Jan 10 17:56:22 2023 +0000
+++ b/hicBuildMatrix.xml	Mon Jul 01 19:04:07 2024 +0000
@@ -5,7 +5,7 @@
         <import>macros.xml</import>
     </macros>
     <expand macro="requirements">
-        <requirement type="package" version="1.9">samtools</requirement>
+        <requirement type="package" version="1.19">samtools</requirement>
     </expand>
     <command detect_errors="exit_code"><![CDATA[
 
@@ -18,7 +18,7 @@
             #end for
 
             --restrictionCutFile '$restrictionCutFile'
-            
+
             #if $restrictionSequence:
                 --restrictionSequence '$restrictionSequence'
             #end if
@@ -31,7 +31,7 @@
             #if $maxLibraryInsertSize:
                 --maxLibraryInsertSize $maxLibraryInsertSize
             #end if
-           
+
             #if $binSizes:
                 --binSize
                 #for $repeat in $binSizes
@@ -39,30 +39,33 @@
                 #end for
             #end if
 
-            --genomeAssembly $samFiles[0].samFile.metadata.dbkey
+            #if $chromosomeSizes:
+                --chromosomeSizes '$chromosomeSizes'
+            #end if
+            #if $dbKey:
+                --genomeAssembly '$dbKey'
+            #else
+                --genomeAssembly '$samFiles[0].samFile.metadata.dbkey'
+            #end if
 
             #if $region:
                 --region '$region'
             #end if
 
-            --outFileName matrix.$outputFormat
+            --outFileName 'matrix.$outputFormat'
 
             #if $outBam:
                 $outBam ./unsorted.bam
             #end if
 
             $keepSelfCircles
-            $removeSelfLigation
+            $keepSelfLigation
             $skipDuplicationCheck
 
             #if $minMappingQuality and $minMappingQuality is not None:
                 --minMappingQuality $minMappingQuality
             #end if
 
-            #if $chromosomeSizes:
-                --chromosomeSizes '$chromosomeSizes'
-            #end if
-
             --threads @THREADS@
 
             --QCfolder ./QCfolder
@@ -81,7 +84,6 @@
         <!-- can we use multiple=True here with min="2" and max="2" ? -->
         <repeat max="2" min="2" name="samFiles" title="Sam/Bam files to process (forward/reverse)" help="Please use the special BAM datatype: qname_input_sorted.bam and use for 'bowtie2' the '--reorder' option to create a BAM file.">
             <param name="samFile" type="data" format="sam,qname_input_sorted.bam">
-                <validator type="unspecified_build" />
             </param>
         </repeat>
 
@@ -96,24 +98,26 @@
                         belong to the same fragment (by checking if they are within this max insert size) and to decide if a mate
                         is too far away from the nearest restriction site." />
 
-        <repeat name='binSizes' title='Bin size in bp' min="1" help="If used, the restriction cut places (if given) are used to only consider reads that are in the vicinity of the resctriction sites.
+        <repeat name="binSizes" title="Bin size in bp" min="1" help="If used, the restriction cut places (if given) are used to only consider reads that are in the vicinity of the resctriction sites.
                 Otherwise all reads in the interval are considered. Use multiple ones to create a mcool file.">
             <param argument="--binSize" type="integer" optional="true" value="" label="Bin size in bp" />
         </repeat>
 
         <expand macro="region" />
         <param argument="--keepSelfCircles" type="boolean" truevalue="--keepSelfCircles" falsevalue="" label="Keep self circles" help="If set, outward facing reads without any restriction fragment (self circles) are kept. They will be counted and shown in the QC plots." />
-        <param argument="--removeSelfLigation" type="boolean" truevalue="--removeSelfLigation" falsevalue="" label="remove self ligation" help="If set, inward facing reads less than 1000 bp apart and having a restriction site in between are removed. Although this reads do not contribute to any distant contact, they are useful to account for bias in the data" />
+        <param argument="--keepSelfLigation" type="boolean" truevalue="--keepSelfLigation" falsevalue="" label="Keep self ligation" help="If set, inward facing reads less than 1000 bp apart and having a restriction site in between are kept. Although this reads do not contribute to any distant contact, they are useful to account for bias in the data." />
         <expand macro="minMappingQuality" />
         <param argument="--skipDuplicationCheck" type="boolean" truevalue="--skipDuplicationCheck" falsevalue="" label="Skip duplication check" help="Identification of duplicated read pairs is memory consuming. Thus, in case of memory errors this check can be skipped." />
         <param argument="--chromosomeSizes" type="data" format="tabular" optional="true" label="Chromosome sizes for your genome" help="File with the chromosome sizes for your genome. A tab-delimited two column layout 'chr_name size' is expected
-                    Usually the sizes can be determined from the SAM/BAM input files, however, 
-                    for cHi-C or scHi-C it can be that at the start or end no data is present. 
+                    Usually the sizes can be determined from the SAM/BAM input files, however,
+                    for cHi-C or scHi-C it can be that at the start or end no data is present.
                     Please consider that this option causes that only reads are considered which are on the listed chromosomes.
-                    Use this option to guarantee fixed sizes. An example file is available via UCSC: 
+                    Use this option to guarantee fixed sizes. An example file is available via UCSC:
                     http://hgdownload.soe.ucsc.edu/goldenPath/dm3/bigZips/dm3.chrom.sizes" />
+        <param name="dbKey" type="text" optional="true" label="Use this dbkey for your history genome"
+        help="You can set the reference genome in your history as metadata. In case you have not you can specify it here." />
 
-        <param argument='--outBam' type='boolean' truevalue='--outBam' falsevalue="" checked="false" label="Save valid Hi-C reads in BAM file" help="A bam
+        <param argument="--outBam" type="boolean" truevalue="--outBam" falsevalue="" checked="false" label="Save valid Hi-C reads in BAM file" help="A bam
                     file containing all valid Hi-C reads can be created
                     using this option. This bam file could be useful to
                     inspect the distribution of valid Hi-C reads pairs or
@@ -121,8 +125,8 @@
                     HiCExplorer tool. Computation will be significantly
                     longer if this option is set." />
 
-        <param name='outputFormat' type='select' label="Output file format">
-            <option value='h5'>HiCExplorer format</option>
+        <param name="outputFormat" type="select" label="Output file format">
+            <option value="h5">HiCExplorer format</option>
             <option value="cool">cool</option>
         </param>
     </inputs>
@@ -136,7 +140,7 @@
             </change_format>
         </data>
         <data name="qc" format="html" label="${tool.name} QC on ${on_string}" />
-        <data name="raw_qc" from_work_dir='raw_qc' format='txt' label="${tool.name} raw QC on ${on_string}" />
+        <data name="raw_qc" from_work_dir="raw_qc" format="txt" label="${tool.name} raw QC on ${on_string}" />
     </outputs>
     <tests>
         <test expect_num_outputs="4">
@@ -146,21 +150,21 @@
             <repeat name="samFiles">
                 <param name="samFile" value="small_test_R2_unsorted.sam" dbkey="hg38" />
             </repeat>
-            <param name='outputFormat' value='h5' />
-            <repeat name='binSizes'>
+            <param name="outputFormat" value="h5" />
+            <repeat name="binSizes">
                 <param name="binSize" value="5000" />
             </repeat>
-            <param name='restrictionCutFile' value='DpnII_10k.bed' />
-            <param name='restrictionSequence' value='GATC' />
-            <param name='danglingSequence' value='GATC' />
-            <param name='outBam' value="True" />
-            <output name="outfileBam" file="small_test_matrix_result_sorted.bam" ftype="bam" />
+            <param name="restrictionCutFile" value="DpnII_10k.bed" />
+            <param name="restrictionSequence" value="GATC" />
+            <param name="danglingSequence" value="GATC" />
+            <param name="outBam" value="True" />
+            <output name="outfileBam" file="small_test_matrix_result_sorted.bam" compare="diff" lines_diff="2" ftype="bam" />
             <output name="outFileName" ftype="h5">
                 <assert_contents>
-                    <has_h5_keys keys='intervals,matrix' />
+                    <has_h5_keys keys="intervals,matrix" />
                 </assert_contents>
             </output>
-            <output name="raw_qc" file='raw_qc_report' compare='diff' lines_diff='2' />
+            <output name="raw_qc" file="raw_qc_report" compare="diff" lines_diff="2" />
         </test>
         <test expect_num_outputs="4">
             <repeat name="samFiles">
@@ -169,21 +173,21 @@
             <repeat name="samFiles">
                 <param name="samFile" value="small_test_R2_unsorted.sam" dbkey="hg38" />
             </repeat>
-            <repeat name='binSizes'>
+            <repeat name="binSizes">
                 <param name="binSize" value="5000" />
             </repeat>
-            <param name='restrictionCutFile' value='DpnII_10k.bed' />
-            <param name='restrictionSequence' value='GATC' />
-            <param name='danglingSequence' value='GATC' />
-            <param name='outputFormat' value='cool' />
-            <param name='outBam' value="True" />
-            <output name="outfileBam" file="small_test_matrix_result_sorted.bam" ftype="bam" />
+            <param name="restrictionCutFile" value="DpnII_10k.bed" />
+            <param name="restrictionSequence" value="GATC" />
+            <param name="danglingSequence" value="GATC" />
+            <param name="outputFormat" value="cool" />
+            <param name="outBam" value="True" />
+            <output name="outfileBam" file="small_test_matrix_result_sorted.bam" compare="diff" lines_diff="2" ftype="bam" />
             <output name="outFileName" ftype="cool">
                 <assert_contents>
-                    <has_h5_keys keys='bins,chroms,indexes,pixels' />
+                    <has_h5_keys keys="bins,chroms,indexes,pixels" />
                 </assert_contents>
             </output>
-            <output name="raw_qc" file='raw_qc_report' compare='diff' lines_diff='2' />
+            <output name="raw_qc" file="raw_qc_report" compare="diff" lines_diff="2" />
         </test>
     </tests>
     <help><![CDATA[