Hello, I run nf-core/chipseq 2.1.0 and found that the Mark Duplicates of the MERGED LIB: Picard (unfiltered) section in the MultiQC seems incorrect. The following H3K27Ac library was sequenced twice and merged. In a results/bowtie2/merged_library/picard_metrics/H3K27Ac_REP1.mLb.mkD.sorted.MarkDuplicates.metrics.txt, the metrics of the technical replicates were written in the two rows separately: ## METRICS CLASS picard.sam.DuplicationMetrics LIBRARY UNPAIRED_READS_EXAMINED READ_PAIRS_EXAMINED SECONDARY_OR_SUPPLEMENTARY_RDS UNMAPPED_READS UNPAIRED_READ_DUPLICATES READ_PAIR_DUPLICATES READ_PAIR_OPTICAL_DUPLICATES PERCENT_DUPLICATION ESTIMATED_LIBRARY_SIZE H3K27Ac_REP1_T1 147989 6164554 0 224851 37578 759418 516670 0.124742 63806670 H3K27Ac_REP1_T2 1333127 44048308 0 1918643 498651 6312181 3683105 0.146741 296264718 However, in the MultiQC yaml file, the metrics of only the technical replicate 1 was collected and displayed in the MultiQC html. H3K27Ac_REP1: Duplicate Pairs Nonoptical: 485496 Duplicate Pairs Optical: 1033340 Duplicate Unpaired: 37578 Unique Pairs: 10810272 Unique Unpaired: 110411 Unmapped: 224851 Is this a bug or specification of the pipeline?