From bca6efcaa923fbd93c05420e95b31c01f3c6e085 Mon Sep 17 00:00:00 2001
From: Harry Hung <4848896+HarryHung@users.noreply.github.com>
Date: Mon, 6 Jan 2025 13:27:28 +0000
Subject: [PATCH] Add thread limitation to samtools sort to avoid excessive
 memory request

---
 bin/convert_sam_to_sorted_bam.sh | 8 ++++++--
 modules/mapping.nf               | 2 ++
 2 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/bin/convert_sam_to_sorted_bam.sh b/bin/convert_sam_to_sorted_bam.sh
index c730a5f..f3278a2 100755
--- a/bin/convert_sam_to_sorted_bam.sh
+++ b/bin/convert_sam_to_sorted_bam.sh
@@ -1,9 +1,13 @@
 # Convet SAM to sorted BAM file
 # Remove source SAM file if $LITE is true
 
-samtools view -@ "$(nproc)" -b "$SAM" > "$BAM"
+# Thread usage is capped as memory usage is per thread, and the speed gain level-off as thread count increases 
+AVAILABLE_THREAD=$(nproc)
+THREAD=$(( AVAILABLE_THREAD > MAX_THREAD ? MAX_THREAD : AVAILABLE_THREAD ))
 
-samtools sort -@ "$(nproc)" -o "$SORTED_BAM" "$BAM"
+samtools view -@ "$THREAD" -b "$SAM" > "$BAM"
+
+samtools sort -@ "$THREAD" -o "$SORTED_BAM" "$BAM"
 rm "$BAM"
 
 if [ "$LITE" = true ]; then
diff --git a/modules/mapping.nf b/modules/mapping.nf
index 08c298b..42f0ace 100644
--- a/modules/mapping.nf
+++ b/modules/mapping.nf
@@ -67,11 +67,13 @@ process SAM_TO_SORTED_BAM {
 
     script:
     sorted_bam="${sample_id}_mapped_sorted.bam"
+    max_thread="8"
     """
     SAM="$sam"
     BAM="mapped.bam"
     SORTED_BAM="$sorted_bam"
     LITE="$lite"
+    MAX_THREAD="$max_thread"
 
     source convert_sam_to_sorted_bam.sh
     source get_ref_coverage.sh