Clustering of 16s rRNA reads in qiime1 and qiime2

User Support
1

I am new to metagenomics analysis and trying to understand qiime1 and qiime2 workflow:

I do have one query related to clustering:

As I am able to understand that in qiime1 the reads which have 97% similarity will be clustered in one OTU and In Qiime2 100% similar reads clustered into one cluster and known as amplicon variant.

For qiime1, I used uclust method and qiime2, I learned that we do have vsearch method.

Please correct me if I am wrong in above statement.

Now my query is 16S rRNA reads may different length, so while clustering do we take long read as seed or the reads which are present in large number.

2

Hey there @Yogesh_Gupta!

You don't have to cluster sequences in QIIME 2 at all --- please see this section of the Overview tutorial for more details: Overview of QIIME 2 Plugin Workflows — QIIME 2 2019.1.0 documentation

In QIIME 2, if you want to cluster, you can use q2-vsearch, as you mentioned, but, the 100% ASVs are the output of DADA2 and deblur, so no need to cluster further. Make sense? :qiime2:

1 Like
3

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.