From 3aaf92fd11a03350f16d18d8644becc654456594 Mon Sep 17 00:00:00 2001
From: Mauro Saporita <64437596+mauro-saporita@users.noreply.github.com>
Date: Mon, 4 Dec 2023 15:58:21 +0100
Subject: [PATCH] Picard extractfingerprint (#4506)

* first commit - Picard picard_extractfingerprint

* add nf-tests

* improved bam list declaration

Co-authored-by: Adam Talbot <12817534+adamrtalbot@users.noreply.github.com>

* change for alphabetical order

Co-authored-by: Adam Talbot <12817534+adamrtalbot@users.noreply.github.com>

* add tags to main.nf.test

* updated snap

* new snap

* picard version 3.1.0

* changed tests

* add test for tbi

* change test solving snapshot issue

* using new nf-test version

* add EOL

* adde new EOL

* correct typo in tags

* version 3.0.0

* revert to version 3.1.1 (latest)

* add linesGzip test

* remove samtools version from main.nf

* fix stub

* fix version.yml

---------

Co-authored-by: Adam Talbot <12817534+adamrtalbot@users.noreply.github.com>
---
 .../picard/extractfingerprint/environment.yml |  7 ++
 .../nf-core/picard/extractfingerprint/main.nf | 66 +++++++++++++++++++
 .../picard/extractfingerprint/meta.yml        | 46 +++++++++++++
 .../extractfingerprint/tests/main.nf.test     | 47 +++++++++++++
 .../tests/main.nf.test.snap                   | 16 +++++
 .../picard/extractfingerprint/tests/tags.yml  |  2 +
 6 files changed, 184 insertions(+)
 create mode 100644 modules/nf-core/picard/extractfingerprint/environment.yml
 create mode 100644 modules/nf-core/picard/extractfingerprint/main.nf
 create mode 100644 modules/nf-core/picard/extractfingerprint/meta.yml
 create mode 100644 modules/nf-core/picard/extractfingerprint/tests/main.nf.test
 create mode 100644 modules/nf-core/picard/extractfingerprint/tests/main.nf.test.snap
 create mode 100644 modules/nf-core/picard/extractfingerprint/tests/tags.yml

diff --git a/modules/nf-core/picard/extractfingerprint/environment.yml b/modules/nf-core/picard/extractfingerprint/environment.yml
new file mode 100644
index 00000000000..40429b416b3
--- /dev/null
+++ b/modules/nf-core/picard/extractfingerprint/environment.yml
@@ -0,0 +1,7 @@
+name: picard_extractfingerprint
+channels:
+  - conda-forge
+  - bioconda
+  - defaults
+dependencies:
+  - bioconda::picard=3.1.1
diff --git a/modules/nf-core/picard/extractfingerprint/main.nf b/modules/nf-core/picard/extractfingerprint/main.nf
new file mode 100644
index 00000000000..c10c3847ead
--- /dev/null
+++ b/modules/nf-core/picard/extractfingerprint/main.nf
@@ -0,0 +1,66 @@
+process PICARD_EXTRACTFINGERPRINT {
+    tag "$meta.id"
+    label 'process_medium'
+
+    conda "${moduleDir}/environment.yml"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'https://depot.galaxyproject.org/singularity/picard:3.1.1--hdfd78af_0' :
+        'biocontainers/picard:3.1.1--hdfd78af_0' }"
+
+    input:
+    tuple val(meta), path(bam), path(bai)
+    path  haplotype_map
+    path  fasta
+    path  fasta_fai
+    path  sequence_dictionary
+
+    output:
+    tuple val(meta), path("*.vcf.gz")    , emit: vcf
+    tuple val(meta), path("*.vcf.gz.tbi"), emit: tbi
+    path "versions.yml"                  , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def args = task.ext.args ?: ''
+    def prefix = task.ext.prefix ?: "${meta.id}"
+
+    def reference = fasta ? "--REFERENCE_SEQUENCE ${fasta}" : ""
+    def bam_name = bam.simpleName
+
+    def avail_mem = 3072
+    if (!task.memory) {
+        log.info '[PICARD ExtractFingerprint] Available memory not known - defaulting to 3GB. Specify process memory requirements to change this.'
+    } else {
+        avail_mem = (task.memory.mega*0.8).intValue()
+    }
+
+    """
+    picard \\
+        -Xmx${avail_mem}M \\
+        ExtractFingerprint \\
+        --INPUT ${bam} \\
+        --HAPLOTYPE_MAP ${haplotype_map} \\
+        --OUTPUT ${prefix}.vcf.gz \\
+        ${reference} \\
+        $args
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        picard: \$(echo \$(picard ExtractFingerprint --version 2>&1) | grep -o 'Version:.*' | cut -f2- -d:)
+    END_VERSIONS
+    """
+
+    stub:
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    """
+    touch ${prefix}.vcf
+    touch ${prefix}.vcf.gz.tbi
+
+    cat <<-END_VERSIONS > versions.yml
+    "${task.process}":
+        picard: \$(echo \$(picard ExtractFingerprint --version 2>&1) | grep -o 'Version:.*' | cut -f2- -d:)
+    END_VERSIONS
+    """
+}
diff --git a/modules/nf-core/picard/extractfingerprint/meta.yml b/modules/nf-core/picard/extractfingerprint/meta.yml
new file mode 100644
index 00000000000..5c6ccb42ccc
--- /dev/null
+++ b/modules/nf-core/picard/extractfingerprint/meta.yml
@@ -0,0 +1,46 @@
+name: "picard_extractfingerprint"
+description: Computes/Extracts the fingerprint genotype likelihoods from the supplied file. It is given as a list of PLs at the fingerprinting sites.
+keywords:
+  - picard
+  - extract
+  - fingerprint
+  - bam
+tools:
+  - picard:
+      description: |
+        A set of command line tools (in Java) for manipulating high-throughput sequencing (HTS)
+        data and formats such as SAM/BAM/CRAM and VCF.
+      homepage: https://broadinstitute.github.io/picard/
+      documentation: https://broadinstitute.github.io/picard/
+      tool_dev_url: https://github.com/broadinstitute/picard/
+      licence: ["MIT"]
+input:
+  - meta:
+      type: file
+      description: |
+        Input SAM/BAM/CRAM file
+  - reference:
+      type: file
+      description: |
+        Reference sequence file
+  - haplotype_map:
+      type: file
+      description: |
+        A file of haplotype information. The file lists a set of SNPs, optionally arranged in high-LD blocks, to be used for fingerprinting.
+        See https://software.broadinstitute.org/gatk/documentation/article?id=9526 for details.
+      pattern: "*.{txt,vcf,vcf.gz}"
+output:
+  - fingerprint:
+      type: file
+      description: |
+        Output fingerprint file (VCF)
+  - versions:
+      type: file
+      description: File containing software versions
+      pattern: "versions.yml"
+authors:
+  - "@adamrtalbot"
+  - "@mauro-saporita"
+maintainers:
+  - "@adamrtalbot"
+  - "@mauro-saporita"
diff --git a/modules/nf-core/picard/extractfingerprint/tests/main.nf.test b/modules/nf-core/picard/extractfingerprint/tests/main.nf.test
new file mode 100644
index 00000000000..13f8534009b
--- /dev/null
+++ b/modules/nf-core/picard/extractfingerprint/tests/main.nf.test
@@ -0,0 +1,47 @@
+nextflow_process {
+
+    name "Test Process PICARD_EXTRACTFINGERPRINT"
+    script "../main.nf"
+    process "PICARD_EXTRACTFINGERPRINT"
+    tag "modules"
+    tag "modules_nfcore"
+    tag "picard"
+    tag "picard/extractfingerprint"
+
+    test("extract_fingerprint") {
+        when {
+            params {
+                outdir   = "$outputDir"
+            }
+            process {
+                """
+                bam = [
+                    [ id:'test', single_end:false ], // meta map
+                    file(params.test_data['homo_sapiens']['illumina']['test_paired_end_sorted_bam'], checkIfExists: true),
+                    file(params.test_data['homo_sapiens']['illumina']['test_paired_end_sorted_bam_bai'], checkIfExists: true)
+                ]
+                fasta = [file(params.test_data['homo_sapiens']['genome']['genome_fasta'], checkIfExists: true)]
+                fai   = [file(params.test_data['homo_sapiens']['genome']['genome_fasta_fai'], checkIfExists: true)]
+                dict  = [file(params.test_data['homo_sapiens']['genome']['genome_dict'], checkIfExists: true)]
+                haplotype_map  = [file(params.test_data['homo_sapiens']['genome']['haplotype_map'], checkIfExists: true)]
+
+                input[0] = bam
+                input[1] = haplotype_map
+                input[2] = fasta
+                input[3] = fai
+                input[4] = dict
+                """
+            }
+        }
+
+        then {
+            assertAll (
+                { assert process.success },
+                { assert process.out.vcf.get(0).get(1) ==~ ".*.vcf.gz" },
+                { assert process.out.tbi.get(0).get(1) ==~ ".*.vcf.gz.tbi" },
+                { assert snapshot(process.out.versions).match("versions") },
+                { assert snapshot(path(process.out.vcf.get(0).get(1)).linesGzip[0]).match() }
+            )
+        }
+    }
+}
diff --git a/modules/nf-core/picard/extractfingerprint/tests/main.nf.test.snap b/modules/nf-core/picard/extractfingerprint/tests/main.nf.test.snap
new file mode 100644
index 00000000000..fe61840bcc6
--- /dev/null
+++ b/modules/nf-core/picard/extractfingerprint/tests/main.nf.test.snap
@@ -0,0 +1,16 @@
+{
+    "versions": {
+        "content": [
+            [
+                "versions.yml:md5,90dca36c10677c9d6ad5ef7e42c4d788"
+            ]
+        ],
+        "timestamp": "2023-12-04T15:53:39.112154"
+    },
+    "extract_fingerprint": {
+        "content": [
+            "##fileformat=VCFv4.2"
+        ],
+        "timestamp": "2023-12-04T15:53:39.12594"
+    }
+}
\ No newline at end of file
diff --git a/modules/nf-core/picard/extractfingerprint/tests/tags.yml b/modules/nf-core/picard/extractfingerprint/tests/tags.yml
new file mode 100644
index 00000000000..5a853126d42
--- /dev/null
+++ b/modules/nf-core/picard/extractfingerprint/tests/tags.yml
@@ -0,0 +1,2 @@
+picard/extractfingerprint:
+  - modules/nf-core/picard/extractfingerprint/**