From fb38a02f2378b3e09264952342c255fc18bf9a99 Mon Sep 17 00:00:00 2001
From: skchronicles <kuhnsa3@gmail.com>
Date: Wed, 7 Feb 2024 11:50:30 -0500
Subject: [PATCH] Use tmpdir/lscratch with fastqc due to gpfs issue

---
 ervx                          |  6 ++--
 resources/cluster.json        |  4 ++-
 workflow/rules/paired-end.smk | 54 +++++++++++++++++++++++++++++++++--
 3 files changed, 58 insertions(+), 6 deletions(-)

diff --git a/ervx b/ervx
index 9917b9f..1594803 100755
--- a/ervx
+++ b/ervx
@@ -1,7 +1,9 @@
 #!/usr/bin/env python
 # -*- coding: UTF-8 -*-
 
-"""RNA-seek: an highly reproducible and portable RNA-seq pipeline
+"""
+This pipeline was bootstrapped from: https://github.com/skchronicles/RNA-seek
+RNA-seek: an highly reproducible and portable RNA-seq pipeline
 About:
     This is the main entry for the RNA-seek pipeline.
 USAGE:
@@ -21,7 +23,7 @@ import argparse  # potential python3 3rd party package, added in python/3.5
 
 # Pipeline Metadata and globals 
 __author__   = 'Skyler Kuhn'
-__version__  = 'v1.9.0'
+__version__  = 'v0.1.0'
 __email__    = 'kuhnsa@nih.gov'
 __home__     =  os.path.dirname(os.path.abspath(__file__))
 _name        = os.path.basename(sys.argv[0])
diff --git a/resources/cluster.json b/resources/cluster.json
index deccf36..da51f93 100644
--- a/resources/cluster.json
+++ b/resources/cluster.json
@@ -21,6 +21,7 @@
 	"fastqc": {
 		"mem": "24g",
 		"threads": "32",
+		"gres": "lscratch:64",
 		"time": "1-00:00:00"
 	},
 	"fastq_screen": {
@@ -56,7 +57,8 @@
 		"time": "1-00:00:00"
 	},
 	"rawfastqc": {
-		"threads": "8"
+		"threads": "8",
+		"gres": "lscratch:64"
 	},
 	"rsem": {
 		"gres": "lscratch:500",
diff --git a/workflow/rules/paired-end.smk b/workflow/rules/paired-end.smk
index 50eab5e..69f28fc 100644
--- a/workflow/rules/paired-end.smk
+++ b/workflow/rules/paired-end.smk
@@ -50,15 +50,40 @@ rule rawfastqc:
     output:
         join(workpath,"rawQC","{name}.R1_fastqc.zip"),
         join(workpath,"rawQC","{name}.R2_fastqc.zip"),
+        tmpdir=tmpdir,
     priority: 2
     params:
         rname='pl_rawfastqc',
         outdir=join(workpath,"rawQC"),
+        tmpdir=tmpdir,
     threads: int(allocated("threads", "rawfastqc", cluster)),
     envmodules: config['bin'][pfamily]['tool_versions']['FASTQCVER']
     container: config['images']['fastqc']
     shell: """
-    fastqc {input.R1} {input.R2} -t {threads} -o {params.outdir};
+    # Setups temporary directory for
+    # intermediate files with built-in 
+    # mechanism for deletion on exit
+    if [ ! -d "{params.tmpdir}" ]; then mkdir -p "{params.tmpdir}"; fi
+    tmp=$(mktemp -d -p "{params.tmpdir}")
+    trap 'rm -rf "${{tmp}}"' EXIT
+
+    # Running fastqc with local
+    # disk or a tmpdir, fastqc
+    # has been observed to lock
+    # up gpfs filesystems, adding
+    # this on request by HPC staff
+    fastqc \\
+        {input.R1} \\
+        {input.R2} \\
+        -t {threads} \\
+        -o "${{tmp}}"
+
+    # Copy output files from tmpdir
+    # to output directory
+    find "${{tmp}}" \\
+        -type f \\
+        \\( -name '*.html' -o -name '*.zip' \\) \\
+        -exec cp {{}} {params.outdir} \\;
     """
 
 
@@ -119,12 +144,35 @@ rule fastqc:
     params:
         rname='pl_fastqc',
         outdir=join(workpath,"QC"),
-        getrl=join("workflow", "scripts", "get_read_length.py"),
+        tmpdir=tmpdir,
     threads: int(allocated("threads", "fastqc", cluster)),
     envmodules: config['bin'][pfamily]['tool_versions']['FASTQCVER']
     container: config['images']['fastqc']
     shell: """
-    fastqc {input.R1} {input.R2} -t {threads} -o {params.outdir};
+    # Setups temporary directory for
+    # intermediate files with built-in 
+    # mechanism for deletion on exit
+    if [ ! -d "{params.tmpdir}" ]; then mkdir -p "{params.tmpdir}"; fi
+    tmp=$(mktemp -d -p "{params.tmpdir}")
+    trap 'rm -rf "${{tmp}}"' EXIT
+
+    # Running fastqc with local
+    # disk or a tmpdir, fastqc
+    # has been observed to lock
+    # up gpfs filesystems, adding
+    # this on request by HPC staff
+    fastqc \\
+        {input.R1} \\
+        {input.R2} \\
+        -t {threads} \\
+        -o "${{tmp}}"
+
+    # Copy output files from tmpdir
+    # to output directory
+    find "${{tmp}}" \\
+        -type f \\
+        \\( -name '*.html' -o -name '*.zip' \\) \\
+        -exec cp {{}} {params.outdir} \\;
     """