how to compare SILVA 138 taxonomy and NCBI taxonomy

Hi there,

I have two datasets, 16S rRNA data that was annotated with SILVA taxonomy, and shotgun data that was annotated with NCBI taxonomy. I want to compare the difference and find common taxa between the two datasets, is there a way to compare them? Thanks!

Hello @Wei_Zhang,

What are the structures of your two datasets? Do you have feature tables for your 16S data? What do you have for your metagenomic data?

Hi @colinvwood, I used UniqKraken to extract and annotate the microbial reads in RNA-Seq data from the tissue sample, which assigned the NCBI taxonomy as UniqKraken adopted the NCBI genomes. For comparison, 16S rRNA sequencing for those tissue samples was also performed, followed by qiime2-DADA2 analysis with SILVA 138 taxonomy. Thus, I have those two feature tables. There are some identical genera as well as different genera. I wonder how to compare them and discover the discrepancy between RNA-Seq and 16S rRNA. Should I only focus on the genera with identical genus names?

Hello @Wei_Zhang,

Are these two datasets from the same extraction or sample? Are you trying to assess the differences introduced by these varying methods? If you're trying to investigate the microbial differences between the two datasets that's going to be difficult given all the variables.

How did you get your shotgun data into feature table format if you don't mind me asking? You could still perform all the typical alpha and beta diversity analyses in qiime. You could merge the separate feature tables before hand.

1 Like