From 31409f5e727ec932f0b3a399c7a3847d70b21374 Mon Sep 17 00:00:00 2001
From: Matthieu Muffato <mm49@sanger.ac.uk>
Date: Tue, 26 Jul 2022 22:13:19 +0100
Subject: [PATCH] Added the samtools/dict module (#1922)

* Added the samtools/dict module

* samtools/dict is single-threaded
---
 modules/samtools/dict/main.nf               | 44 +++++++++++++++++++++
 modules/samtools/dict/meta.yml              | 41 +++++++++++++++++++
 tests/config/pytest_modules.yml             |  4 ++
 tests/modules/samtools/dict/main.nf         | 13 ++++++
 tests/modules/samtools/dict/nextflow.config |  5 +++
 tests/modules/samtools/dict/test.yml        | 13 ++++++
 6 files changed, 120 insertions(+)
 create mode 100644 modules/samtools/dict/main.nf
 create mode 100644 modules/samtools/dict/meta.yml
 create mode 100644 tests/modules/samtools/dict/main.nf
 create mode 100644 tests/modules/samtools/dict/nextflow.config
 create mode 100644 tests/modules/samtools/dict/test.yml

diff --git a/modules/samtools/dict/main.nf b/modules/samtools/dict/main.nf
new file mode 100644
index 00000000..91f782b1
--- /dev/null
+++ b/modules/samtools/dict/main.nf
@@ -0,0 +1,44 @@
+process SAMTOOLS_DICT {
+    tag "$fasta"
+    label 'process_single'
+
+    conda (params.enable_conda ? "bioconda::samtools=1.15.1" : null)
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/samtools:1.15.1--h1170115_0' :
+        'quay.io/biocontainers/samtools:1.15.1--h1170115_0' }"
+
+    input:
+    tuple val(meta), path(fasta)
+
+    output:
+    tuple val(meta), path ("*.dict"), emit: dict
+    path "versions.yml"             , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args = task.ext.args ?: ''
+    """
+    samtools \\
+        dict \\
+        $args \\
+        $fasta \\
+        > ${fasta}.dict
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        samtools: \$(echo \$(samtools --version 2>&1) | sed 's/^.*samtools //; s/Using.*\$//')
+    END_VERSIONS
+    """
+
+    stub:
+    """
+    touch ${fasta}.dict
+    cat <<-END_VERSIONS > versions.yml
+
+    "${task.process}":
+        samtools: \$(echo \$(samtools --version 2>&1) | sed 's/^.*samtools //; s/Using.*\$//')
+    END_VERSIONS
+    """
+}
diff --git a/modules/samtools/dict/meta.yml b/modules/samtools/dict/meta.yml
new file mode 100644
index 00000000..e3eeccc8
--- /dev/null
+++ b/modules/samtools/dict/meta.yml
@@ -0,0 +1,41 @@
+name: samtools_dict
+description: Create a sequence dictionary file from a FASTA file
+keywords:
+  - dict
+  - fasta
+tools:
+  - samtools:
+      description: |
+        SAMtools is a set of utilities for interacting with and post-processing
+        short DNA sequence read alignments in the SAM, BAM and CRAM formats, written by Heng Li.
+        These files are generated as output by short read aligners like BWA.
+      homepage: http://www.htslib.org/
+      documentation: http://www.htslib.org/doc/samtools.html
+      doi: 10.1093/bioinformatics/btp352
+      licence: ["MIT"]
+input:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test', single_end:false ]
+  - fasta:
+      type: file
+      description: FASTA file
+      pattern: "*.{fa,fasta}"
+output:
+  - meta:
+      type: map
+      description: |
+        Groovy Map containing sample information
+        e.g. [ id:'test', single_end:false ]
+  - dict:
+      type: file
+      description: FASTA dictionary file
+      pattern: "*.{dict}"
+  - versions:
+      type: file
+      description: File containing software versions
+      pattern: "versions.yml"
+authors:
+  - "@muffato"
diff --git a/tests/config/pytest_modules.yml b/tests/config/pytest_modules.yml
index b7f84004..bcbee2b4 100644
--- a/tests/config/pytest_modules.yml
+++ b/tests/config/pytest_modules.yml
@@ -1919,6 +1919,10 @@ samtools/depth:
   - modules/samtools/depth/**
   - tests/modules/samtools/depth/**
 
+samtools/dict:
+  - modules/samtools/dict/**
+  - tests/modules/samtools/dict/**
+
 samtools/faidx:
   - modules/samtools/faidx/**
   - tests/modules/samtools/faidx/**
diff --git a/tests/modules/samtools/dict/main.nf b/tests/modules/samtools/dict/main.nf
new file mode 100644
index 00000000..f24a8382
--- /dev/null
+++ b/tests/modules/samtools/dict/main.nf
@@ -0,0 +1,13 @@
+#!/usr/bin/env nextflow
+
+nextflow.enable.dsl = 2
+
+include { SAMTOOLS_DICT } from '../../../../modules/samtools/dict/main.nf'
+
+workflow test_samtools_dict {
+
+    input = [ [ id:'test', single_end:false ], // meta map
+              file(params.test_data['sarscov2']['genome']['genome_fasta'], checkIfExists: true) ]
+
+    SAMTOOLS_DICT ( input )
+}
diff --git a/tests/modules/samtools/dict/nextflow.config b/tests/modules/samtools/dict/nextflow.config
new file mode 100644
index 00000000..8730f1c4
--- /dev/null
+++ b/tests/modules/samtools/dict/nextflow.config
@@ -0,0 +1,5 @@
+process {
+
+    publishDir = { "${params.outdir}/${task.process.tokenize(':')[-1].tokenize('_')[0].toLowerCase()}" }
+
+}
diff --git a/tests/modules/samtools/dict/test.yml b/tests/modules/samtools/dict/test.yml
new file mode 100644
index 00000000..8a71398a
--- /dev/null
+++ b/tests/modules/samtools/dict/test.yml
@@ -0,0 +1,13 @@
+- name: samtools dict test_samtools_dict
+  command: nextflow run tests/modules/samtools/dict -entry test_samtools_dict -c tests/config/nextflow.config -c tests/modules/samtools/dict/nextflow.config
+  tags:
+    - samtools
+    - samtools/dict
+  files:
+    - path: output/samtools/genome.fasta.dict
+      contains:
+        - "SN:MT192765.1"
+        - "LN:29829"
+        - "M5:c95f3e5592d0ad9974e41e7f0ea14eb0"
+    - path: output/samtools/versions.yml
+      md5sum: 12cd0c5ce466eefb2dff72625ecbe5c2