Update STITCH

modules/nf-core/stitch/environment.yml

@@ -4,7 +4,7 @@ channels:
   - conda-forge
   - bioconda
 dependencies:
-  - bioconda::htslib=1.18
-  - bioconda::r-stitch=1.6.10
-  - conda-forge::r-base=4.3.1
-  - conda-forge::rsync=3.2.7
+  - bioconda::htslib=1.21
+  - bioconda::r-stitch=1.7.3
+  - conda-forge::r-base=4.4.2
+  - conda-forge::rsync=3.4.1

modules/nf-core/stitch/main.nf

@@ -4,14 +4,13 @@ process STITCH {
 
     conda "${moduleDir}/environment.yml"
     container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
-        'https://depot.galaxyproject.org/singularity/r-stitch:1.6.10--r43h06b5641_0':
-        'biocontainers/r-stitch:1.6.10--r43h06b5641_0' }"
+        'https://depot.galaxyproject.org/singularity/r-stitch:1.7.3--r44h64f727c_0':
+        'biocontainers/r-stitch:1.7.3--r44h64f727c_0' }"
 
     input:
-    tuple val(meta), path(collected_crams), path(collected_crais), path(cramlist)
-    tuple val(meta2), path(posfile), path(input, stageAs: "input"), path(rdata, stageAs: "RData_in"), val(chromosome_name), val(K), val(nGen)
-    tuple val(meta3), path(fasta), path(fasta_fai)
-    val seed
+    tuple val(meta), path(collected_crams), path(collected_crais), path(cramlist), path(samplename), path(posfile), path(input, stageAs: "input"), path(rdata, stageAs: "RData_in"), val(chromosome_name), val(K), val(nGen)
+    tuple val(meta2), path(fasta), path(fasta_fai)
+    val(seed)
 
     output:
     tuple val(meta), path("input", type: "dir") , emit: input
@@ -30,6 +29,7 @@ process STITCH {
     def args2                = task.ext.args2  ?: ""
     def generate_input_only  = args2.contains( "--generateInputOnly TRUE" )
     def bgen_output          = args2.contains( "--output_format bgen" )
+    def suffix               = bgen_output     ? "bgen"    : "vcf.gz"
     def reads_ext            = collected_crams             ? collected_crams.extension.unique()                                : []
     def rsync_cmd            = rdata                       ? "rsync -rL ${rdata}/ RData"                                       : ""
     def stitch_cmd           = seed                        ? "Rscript <(cat \$(which STITCH.R) | tail -n +2 | cat <(echo 'set.seed(${seed})') -)" : "STITCH.R"
@@ -38,6 +38,7 @@ process STITCH {
     def reference_cmd        = fasta                       ? "--reference ${fasta}"                                            : ""
     def regenerate_input_cmd = input && rdata && !cramlist ? "--regenerateInput FALSE --originalRegionName ${chromosome_name}" : ""
     def rsync_version_cmd    = rdata                       ? "rsync: \$(rsync --version | head -n1 | sed 's/^rsync  version //; s/ .*\$//')" : ""
+    def samplename_cmd       = samplename                  ? "--sampleNames_file ${samplename}"                               : ""
     """
     ${rsync_cmd} ${args}
 
@@ -52,6 +53,8 @@ process STITCH {
         ${bamlist_cmd} \\
         ${reference_cmd} \\
         ${regenerate_input_cmd} \\
+        ${samplename_cmd} \\
+        --output_filename ${prefix}.${suffix} \\
         ${args2}
 
     cat <<-END_VERSIONS > versions.yml
@@ -64,17 +67,41 @@ process STITCH {
 
     stub:
     def prefix               = task.ext.prefix      ?: "${meta.id}"
-    def args                 = task.ext.args        ?: ""
+    def _args                = task.ext.args        ?: ""
     def args2                = task.ext.args2       ?: ""
+    def nb_samples           = collected_crams.size()
     def generate_input_only  = args2.contains( "--generateInputOnly TRUE" )
-    def generate_plots_cmd   = !generate_input_only ? "mkdir plots"                                                                   : ""
-    def generate_vcf_cmd     = !generate_input_only ? "touch ${prefix}.vcf.gz"                                                        : ""
+    def bgen_output          = args2.contains( "--output_format bgen" )
+    def generate_plots_cmd   = !generate_input_only
+    def generate_file_cmd    = !generate_input_only ? bgen_output ? "touch ${prefix}.bgen" : "echo '' | gzip > ${prefix}.vcf.gz"            : ""
     def rsync_version_cmd    = rdata                ? "rsync: \$(rsync --version | head -n1 | sed 's/^rsync  version //; s/ .*\$//')" : ""
     """
-    touch input
-    touch RData
-    ${generate_plots_cmd}
-    ${generate_vcf_cmd}
+    mkdir -p input
+    for i in {1..$nb_samples}
+    do
+        touch "input/sample.\$i.input.${chromosome_name}.RData"
+    done
+
+    ${generate_file_cmd}
+
+    mkdir -p RData
+    touch "RData/EM.all.${chromosome_name}.RData"
+    touch "RData/end.${chromosome_name}.RData"
+    touch "RData/sampleNames.${chromosome_name}.RData"
+    touch "RData/start.${chromosome_name}.RData"
+    touch "RData/startEM.${chromosome_name}.RData"
+
+    if [ "${generate_plots_cmd}" == true ]
+    then
+        mkdir -p plots
+        touch "plots/alphaMat.${chromosome_name}.all.s.1.png"
+        touch "plots/alphaMat.${chromosome_name}.normalized.s.1.png"
+        touch "plots/hapSum.${chromosome_name}.s.1.png"
+        touch "plots/hapSum_log.${chromosome_name}.s.1.png"
+        touch "plots/metricsForPostImputationQC.${chromosome_name}.sample.jpg"
+        touch "plots/metricsForPostImputationQCChromosomeWide.${chromosome_name}.sample.jpg"
+        touch "plots/r2.${chromosome_name}.goodonly.jpg"
+    fi
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":

modules/nf-core/stitch/meta.yml

@@ -1,4 +1,3 @@
-# yaml-language-server: $schema=https://raw.githubusercontent.com/nf-core/modules/master/modules/yaml-schema.json
 name: "stitch"
 description: "STITCH is an R program for reference panel free, read aware, low coverage
   sequencing genotype imputation. STITCH runs on a set of samples with sequencing
@@ -40,11 +39,11 @@ input:
         description: |
           Text file with the path to the cram files to use in imputation, one per line. Since the cram files are staged to the working directory for the process, this file should just contain the file names without any pre-pending path.
         pattern: "*.txt"
-  - - meta2:
-        type: map
-        description: |
-          Groovy Map containing information about the set of positions to run the imputation over
-          e.g. `[ id:'test' ]`
+    - samplename:
+        type: file
+        description: (Optional) File with list of samples names in the same order as in bamlist to impute.
+          One file per line.
+        pattern: "*.{txt}"
     - posfile:
         type: file
         description: |
@@ -73,7 +72,7 @@ input:
         description: Number of generations since founding of the population to use for
           imputation. Refer to the documentation for the `--nGen` argument of STITCH
           for more information.
-  - - meta3:
+  - - meta2:
         type: map
         description: |
           Groovy Map containing information about the reference genome used