Qiime version and singletons

User Support
,
1

Hello everyone, I’m running into a difference between QIIME 2 versions and I’d like to understand what’s going on.

At the end of 2025, I performed 16S analyses using QIIME 2 version 2025.10.1. With that version, the feature table generated by DADA2 still contained some singletons.

A few days ago I switched to QIIME 2 version 2026.4. Compared to a colleague of mine who is still using an older version (2024.5), we now get different results from the same dataset: he still obtains some singletons in the feature table, while I do not.

the command we are using is the following:
"qiime dada2 denoise-paired --i-demultiplexed-seqs demuxtrimmed.qza --p-trunc-len-f 0 --p-trunc-len-r 220 --p-max-ee-f 2 --p-max-ee-r 2 --o-representative-sequences asv-seqs.qza --o-table asv-table.qza --o-base-transition-stats basetransstat.qza --o-denoising-stats statsdada.qza"

Is it possible that newer versions of QIIME 2 (and therefore DADA2) use updated algorithms or filtering steps that tend to remove singleton sequences more aggressively compared to older versions?

Thank you in advance for any clarification.

3

@Alessandro - would it be possible for you to provide the two feature tables that you're comparing? We'd like to compare the data provenance to start exploring what might be going on. Replying to this post with them attached is easiest, but we could also share by DM if you don't want to post the feature tables publicly on the forum. Thanks!

4

Hello, after looking into the issue more closely, it turns out there was a misunderstanding. He actually obtains ASVs from QIIME just like I do, but then generates a final feature table after performing additional analyses outside of QIIME. That’s why our results are slightly different. So for now, we’ll leave it at that, as the differences are most likely due to these different analytical approaches. I apologize for having written the post before investigating more thoroughly, and I thank everyone for their availability!

1 Like
5

Not a problem - glad it's a false alarm. Good luck with your project!