during the analysis of V3 region of 16S Next gene consensus. should I remove denovo chimera or proceed for clustering (OTU formation) without chimera?
what should be the minimum % of identity (90, 95, 97) for clustering?
Chimera removal is must recommended for PCR based community analysis and 97% is the most accepted cutoff level for OTU clustering (for Bacteria and Archaea). However, you can also try multiple cutoff levels for OTU clustering (eg. 95, 97 and 99).
I would say denovo chimera removal is a must. Although you can use a second approch to reference based chimera removal which I generally do.. As suggested the reference is generally used with GOLD databse.
I have certain concerns with the cutoff.. As is discussed earlier, when you do Sanger sequencing a 97% cutoff seems legit, but remember we are dealing with much lower fragment size, any of the V1-V9 or combination, so scaling down the size and applying the same cutoff is not recommended. Having said that, NGS amplicon data still use 97% cutoff by the research community or you can use multiple cutoff levels for OTU clustering as suggested above.
But singleton removal is a must!!!!
- Badges
- Science topic
- Similar topics
- Biological Science
- Genetics
- Genetic Analysis