[FIX] Correct logic for linear deduplication

CHANGELOG.md

@@ -3,6 +3,12 @@
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/)
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [3.2.3] - 2025-08-22
+
+### Fixes
+- Fixed logic error in linear deduplication where the reads with the lowest MAPQ were kept instead.
+
+
 ## [3.2.2] - 2024-02-01
 
 ### Enhancements

bin/find_unique_reads.py

@@ -88,7 +88,7 @@
             alignments[unique_id] = [row[3], row[4]]
 
         # If ID already exists, check whether mapping quality is better
-        elif alignments[unique_id][1] > row[4]:
+        elif alignments[unique_id][1] < row[4]:
             alignments[unique_id] = [row[3], row[4]]
 
         # Increase counter
@@ -98,12 +98,12 @@
 alignments = [i[1][0][:-2] for i in alignments.items()]
 
 # Line to be saved in MultiQC file
-mqc_line = f"LA duplicates removed (%)\t{round((i-len(alignments))/i*100, 2)}"
+mqc_line = f"LA duplicates removed (%)\t{round((i - len(alignments)) / i * 100, 2) if i > 0 else 'NA'}"
 
 # Collect metrics into a string
 report = "LINEAR AMPLIFICATION DUPLICATION METRICS"
 report += f"\nReads before filtering\t{i}"
-report += f"\nLA duplicates removed (n)\t{i-len(alignments)}"
+report += f"\nLA duplicates removed (n)\t{i - len(alignments)}"
 report += "\n" + mqc_line
 report += f"\nUnique reads after LA duplicate removal\t{len(alignments)}"
 

nextflow.config

@@ -338,7 +338,7 @@ manifest {
     description     = """Analysis pipeline for CUT&RUN and CUT&TAG experiments that includes sequencing QC, spike-in normalisation, IgG control normalisation, peak calling and downstream peak analysis."""
     mainScript      = 'main.nf'
     nextflowVersion = '!>=23.04.0'
-    version         = '3.2.2'
+    version         = '3.2.3'
     doi             = 'https://doi.org/10.5281/zenodo.5653535'
 }