New module: Ultra (#871)

* 👌 IMPROVE: Update .gitignore

* 📦 Add ultra module

* 👌 IMPROVE: Update test input

* 👌 IMPROVE: Update and clean code

- Update to last versions.yml file
- Update meta.yml
- Correct typos

* 👌 IMPROVE: Update output channels + Rename following subtool

* 👌 IMPROVE: Remove old ultre files

* 👌 IMPROVE: Update of pytest_modules.yml

* 👌 IMPROVE: Update test.yml

* 👌 IMPROVE: Keep md5sum as much as possible

* 👌 IMPROVE: Remove old ultra files

* 👌 IMPROVE: Update of pytest_modules.yml

* 👌 IMPROVE: Update test.yml

* 👌 IMPROVE: Keep md5sum as much as possible

* 🐛 Fix: add unsaved modifications

* 🐛 FIX: Remove one inconstant md5sum

* 🐛 FIX: Grab software name using ${getSoftwareName(task.process)}

* 🐛 FIX: Remove md5sums for pickle files (not constant).

* Update modules/ultra/pipeline/main.nf

Co-authored-by: Harshil Patel <drpatelh@users.noreply.github.com>

* Update modules/ultra/pipeline/main.nf

Co-authored-by: Harshil Patel <drpatelh@users.noreply.github.com>

* 👌 IMPROVE: update output directory, update meta.yml

* 👌 IMPROVE: Use modules to gunzip and sort gtf

* 🐛 FIX: Set up channel correctly

* 👌 IMPROVE: Remove pickles files and databases

Those data might be useful in a debugging purpose.

* Apply suggestions from code review

* Update main.nf

Co-authored-by: Harshil Patel <drpatelh@users.noreply.github.com>

modules/ultra/pipeline/functions.nf

@@ -0,0 +1,78 @@
+//
+//  Utility functions used in nf-core DSL2 module files
+//
+
+//
+// Extract name of software tool from process name using $task.process
+//
+def getSoftwareName(task_process) {
+    return task_process.tokenize(':')[-1].tokenize('_')[0].toLowerCase()
+}
+
+//
+// Extract name of module from process name using $task.process
+//
+def getProcessName(task_process) {
+    return task_process.tokenize(':')[-1]
+}
+
+//
+// Function to initialise default values and to generate a Groovy Map of available options for nf-core modules
+//
+def initOptions(Map args) {
+    def Map options = [:]
+    options.args            = args.args ?: ''
+    options.args2           = args.args2 ?: ''
+    options.args3           = args.args3 ?: ''
+    options.publish_by_meta = args.publish_by_meta ?: []
+    options.publish_dir     = args.publish_dir ?: ''
+    options.publish_files   = args.publish_files
+    options.suffix          = args.suffix ?: ''
+    return options
+}
+
+//
+// Tidy up and join elements of a list to return a path string
+//
+def getPathFromList(path_list) {
+    def paths = path_list.findAll { item -> !item?.trim().isEmpty() }      // Remove empty entries
+    paths     = paths.collect { it.trim().replaceAll("^[/]+|[/]+\$", "") } // Trim whitespace and trailing slashes
+    return paths.join('/')
+}
+
+//
+// Function to save/publish module results
+//
+def saveFiles(Map args) {
+    def ioptions  = initOptions(args.options)
+    def path_list = [ ioptions.publish_dir ?: args.publish_dir ]
+
+    // Do not publish versions.yml unless running from pytest workflow
+    if (args.filename.equals('versions.yml') && !System.getenv("NF_CORE_MODULES_TEST")) {
+        return null
+    }
+    if (ioptions.publish_by_meta) {
+        def key_list = ioptions.publish_by_meta instanceof List ? ioptions.publish_by_meta : args.publish_by_meta
+        for (key in key_list) {
+            if (args.meta && key instanceof String) {
+                def path = key
+                if (args.meta.containsKey(key)) {
+                    path = args.meta[key] instanceof Boolean ? "${key}_${args.meta[key]}".toString() : args.meta[key]
+                }
+                path = path instanceof String ? path : ''
+                path_list.add(path)
+            }
+        }
+    }
+    if (ioptions.publish_files instanceof Map) {
+        for (ext in ioptions.publish_files) {
+            if (args.filename.endsWith(ext.key)) {
+                def ext_list = path_list.collect()
+                ext_list.add(ext.value)
+                return "${getPathFromList(ext_list)}/$args.filename"
+            }
+        }
+    } else if (ioptions.publish_files == null) {
+        return "${getPathFromList(path_list)}/$args.filename"
+    }
+}

modules/ultra/pipeline/main.nf

@@ -0,0 +1,48 @@
+// Import generic module functions
+include { initOptions; saveFiles; getSoftwareName; getProcessName } from './functions'
+
+params.options = [:]
+options        = initOptions(params.options)
+
+process ULTRA_PIPELINE {
+    tag "$meta.id"
+    label 'process_high'
+    publishDir "${params.outdir}",
+        mode: params.publish_dir_mode,
+        saveAs: { filename -> saveFiles(filename:filename, options:params.options, publish_dir:getSoftwareName(task.process), meta:meta, publish_by_meta:['id']) }
+
+    conda (params.enable_conda ? "bioconda::ultra_bioinformatics=0.0.4" : null)
+    if (workflow.containerEngine == 'singularity' && !params.singularity_pull_docker_container) {
+        container "https://depot.galaxyproject.org/singularity/ultra_bioinformatics:0.0.4--pyh5e36f6f_1"
+    } else {
+        container "quay.io/biocontainers/ultra_bioinformatics:0.0.4--pyh5e36f6f_1"
+    }
+
+    input:
+    tuple val(meta), path(reads)
+    path genome
+    path gtf
+
+    output:
+    tuple val(meta), path("*.sam"), emit: sam
+    path "versions.yml"           , emit: versions
+
+    script:
+    def prefix = options.suffix ? "${meta.id}${options.suffix}" : "${meta.id}"
+    """
+    uLTRA \\
+        pipeline \\
+        --t $task.cpus \\
+        --prefix $prefix \\
+        $options.args \\
+        \$(pwd)/$genome \\
+        \$(pwd)/$gtf \\
+        \$(pwd)/$reads \\
+        ./
+
+    cat <<-END_VERSIONS > versions.yml
+    ${getProcessName(task.process)}:
+        ${getSoftwareName(task.process)}: \$( uLTRA --version|sed 's/uLTRA //g' )
+    END_VERSIONS
+    """
+}

modules/ultra/pipeline/meta.yml

@@ -0,0 +1,50 @@
+name: ultra_pipeline
+description: uLTRA aligner - A wrapper around minimap2 to improve small exon detection
+keywords:
+  - uLTRA
+  - minimap2
+tools:
+  - ultra:
+      description: Splice aligner of long transcriptomic reads to genome.
+      homepage: https://github.com/ksahlin/uLTRA
+      documentation: https://github.com/ksahlin/uLTRA
+      tool_dev_url: https://github.com/ksahlin/uLTRA
+      doi: "10.1093/bioinformatics/btab540"
+      licence: ['GNU GPLV3']
+
+input:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test', single_end:false ]
+  - genome:
+      type: file
+      description: fasta file of reference genome
+      pattern: "*.fasta"
+  - gtf:
+      type: file
+      description: A annotation of use the genome
+      pattern: "*.gtf"
+  - reads:
+      type: file
+      description: A fasta or fastq file of reads to align
+      pattern: "*.{fasta,fastq}"
+
+output:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test', single_end:false ]
+  - sam:
+      type: file
+      description: The aligned reads in sam format
+      pattern: "*.sam"
+  - versions:
+      type: file
+      description: File containing software versions
+      pattern: "versions.yml"
+
+authors:
+  - "@sguizard"

tests/config/pytest_modules.yml

@@ -1311,6 +1311,10 @@ ucsc/wigtobigwig:
   - modules/ucsc/wigtobigwig/**
   - tests/modules/ucsc/wigtobigwig/**
 
+ultra/pipeline:
+  - modules/ultra/pipeline/**
+  - tests/modules/ultra/pipeline/**
+
 unicycler:
   - modules/unicycler/**
   - tests/modules/unicycler/**

tests/modules/ultra/pipeline/main.nf

@@ -0,0 +1,25 @@
+#!/usr/bin/env nextflow
+
+nextflow.enable.dsl = 2
+
+include { ULTRA_PIPELINE } from '../../../../modules/ultra/pipeline/main.nf' addParams( options: [:] )
+include { GUNZIP         } from '../../../../modules/gunzip/main.nf'         addParams( options: [:] )
+include { GFFREAD        } from '../../../../modules/gffread/main.nf'        addParams( options: [args: "--sort-alpha --keep-genes -T", suffix: "_sorted"] )
+
+workflow test_ultra_pipeline {
+
+    fastq  = file(params.test_data['homo_sapiens']['pacbio']['hifi']        , checkIfExists: true)
+    gtf    = file(params.test_data['homo_sapiens']['genome']['genome_gtf']  , checkIfExists: true)
+    genome = file(params.test_data['homo_sapiens']['genome']['genome_fasta'], checkIfExists: true)
+
+    GUNZIP ( fastq )
+    GFFREAD ( gtf )
+
+    GUNZIP
+        .out
+        .gunzip
+        .map { [ [ id:'test', single_end:false ], it ] }
+        .set { input }
+
+    ULTRA_PIPELINE ( input, genome, GFFREAD.out.gtf )
+}

tests/modules/ultra/pipeline/test.yml

@@ -0,0 +1,12 @@
+- name: ultra pipeline test_ultra_pipeline
+  command: nextflow run tests/modules/ultra/pipeline -entry test_ultra_pipeline -c tests/config/nextflow.config
+  tags:
+    - ultra/pipeline
+    - ultra
+  files:
+    - path: output/gffread/genome_sorted.gtf
+      md5sum: c0b034860c679a354cd093109ed90437
+    - path: output/gunzip/test_hifi.fastq
+      md5sum: 20e41c569d5828c1e87337e13a5185d3
+    - path: output/ultra/test.sam
+      md5sum: a37a1f9594a3099522dc1f6a903b2b12