High number of "orphan consensus"

[wiedenhoeft]

I am trying to call UMI consensus using

fgbio -XX:-UseGCOverheadLimit -Xms750m -Xmx8g
--tmp-dir=/tmpfs
CallMolecularConsensusReads
--min-reads=1
--min-input-base-quality=0
--max-reads=1000000
--output-per-base-tags=true
--sort-order=:none:
--debug=true
--input=/dev/stdin
--output=${output_bam}
--rejects=${output_rejects}
2> ${log}

However, a large number of reads get filtered out "due to orphan consensus created".

[2024/11/22 15:59:21 | CallMolecularConsensusReads | Info] Total Raw Reads Considered: 8,314,392.
[2024/11/22 15:59:21 | CallMolecularConsensusReads | Info] Raw Reads Filtered Due to Orphan Consensus Created: 2,356,267 (0.2834).
[2024/11/22 15:59:21 | CallMolecularConsensusReads | Info] Raw Reads Filtered Due to Mismatching Cigars: 27,582 (0.0033).
[2024/11/22 15:59:21 | CallMolecularConsensusReads | Info] Raw Reads Filtered Due to Insufficient Support: 0 (0.0000).
[2024/11/22 15:59:21 | CallMolecularConsensusReads | Info] Consensus reads emitted: 1,906,142.

Previous steps included creation of MI tags and sorting by TemplateCoordinate as usual.

I'm not sure I get what is going on here. What could be the reason for this, and how exactly is "orphan consensus"defined?

Thanks :-)