Regarding the ASV_ID in the ASV feature table

Hello teachers,

I have a question for you regarding my analysis of 16S amplicon data, which I divided into two groups. Each group has its own ASV feature table generated using QIIME2 (2023.09). In my subsequent analysis, I plan to merge these two ASV feature tables into one. Upon reviewing the two feature tables, I noticed that some ASV_IDs are duplicated. Therefore, I would like to ask whether the same ASV_ID (for example, 4f990663f0e481d4bd28bf70a664f8e6) in both tables represents the same ASV.

In other words, is this ASV_ID randomly generated, or does it correspond to a fixed ASV sequence?

Looking forward to your response

Hello!

ASV IDs are Md5 hashes of the sequence itself, so it is not randomly generated and based on the sequence. The same sequences from different tables / runs will have the same IDs.

If you are planning to merge two different runs, make sure that you run all steps prior Dada2 and Dada2 itself with absolutely identical parameters (for example, exactly same commands for primers removal, exactly the same settings in Dada2, like trimming, truncation etc).

Best,

2 Likes

Thank you very much for your assistance. It was incredibly helpful!