From 9ba6b02bbcb322ff4265cc51fca0ee5d8420b929 Mon Sep 17 00:00:00 2001
From: Adam Talbot <12817534+adamrtalbot@users.noreply.github.com>
Date: Wed, 7 Aug 2024 09:19:54 +0100
Subject: [PATCH] Replace bedtools sort with unix sort in BEDTOOLS_GENOMECOV
 (#6063)

* Replace bedtools sort with unix sort in BEDTOOLS_GENOMECOV

`bedtools sort` uses a large amount of CPUs and memory, but when using it here it doesn't require the  additional genome based features of `bedtools`. Replacing it should speed up the process and make it many times more efficient.

* add args2 for for customisation of GNU sort command

Allows customisation of GNU

* quoting for args2

* Use LC_ALL and default options for performance and consistency

* Handle null memory value

* Remove tags.yml
---
 modules/nf-core/bedtools/genomecov/main.nf        | 11 +++++++----
 modules/nf-core/bedtools/genomecov/tests/tags.yml |  2 --
 2 files changed, 7 insertions(+), 6 deletions(-)
 delete mode 100644 modules/nf-core/bedtools/genomecov/tests/tags.yml

diff --git a/modules/nf-core/bedtools/genomecov/main.nf b/modules/nf-core/bedtools/genomecov/main.nf
index 954e8913d3b..8403c530380 100644
--- a/modules/nf-core/bedtools/genomecov/main.nf
+++ b/modules/nf-core/bedtools/genomecov/main.nf
@@ -4,8 +4,8 @@ process BEDTOOLS_GENOMECOV {
 
     conda "${moduleDir}/environment.yml"
     container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
-        'https://depot.galaxyproject.org/singularity/bedtools:2.31.1--hf5e1c6e_0' :
-        'biocontainers/bedtools:2.31.1--hf5e1c6e_0' }"
+        'oras://community.wave.seqera.io/library/bedtools_coreutils:ba273c06a3909a15':
+        'community.wave.seqera.io/library/bedtools_coreutils:a623c13f66d5262b' }"
 
     input:
     tuple val(meta), path(intervals), val(scale)
@@ -21,13 +21,16 @@ process BEDTOOLS_GENOMECOV {
     task.ext.when == null || task.ext.when
 
     script:
-    def args = task.ext.args ?: ''
+    def args      = task.ext.args  ?: ''
     def args_list = args.tokenize()
     args += (scale > 0 && scale != 1) ? " -scale $scale" : ""
     if (!args_list.contains('-bg') && (scale > 0 && scale != 1)) {
         args += " -bg"
     }
-    def sort_cmd = sort ? '| bedtools sort' : ''
+    // Sorts output file by chromosome and position using additional options for performance and consistency
+    // See https://www.biostars.org/p/66927/ for further details
+    def buffer   = task.memory ? "--buffer-size=${task.memory.toGiga().intdiv(2)}G" : ''
+    def sort_cmd = sort ? "| LC_ALL=C sort --parallel=$task.cpus $buffer -k1,1 -k2,2n" : ''
 
     def prefix = task.ext.prefix ?: "${meta.id}"
     if (intervals.name =~ /\.bam/) {
diff --git a/modules/nf-core/bedtools/genomecov/tests/tags.yml b/modules/nf-core/bedtools/genomecov/tests/tags.yml
deleted file mode 100644
index 55fce47800f..00000000000
--- a/modules/nf-core/bedtools/genomecov/tests/tags.yml
+++ /dev/null
@@ -1,2 +0,0 @@
-bedtools/genomecov:
-  - "modules/nf-core/bedtools/genomecov/**"