50% unassigned sequences after taxonomy

[Nicholas_Bokulich]

Welcome @Lu_Wang!

@Lu_Wang

I have tried to use my own classifiers to assign the rep-seqs under Qiime2, but the ratio of “unassigned” sequences is still about 50%, which can not be assigned to kingdom.

From your description, it sounds like you are aware that the mixed orientation of your reads (mix of forward and reverse sequences) is causing the 50% unassigned rate. This is because the classify-sklearn method can only handle reads in a single orientation. They can be forward or reverse relative to the reference, but all query sequences must be oriented.

@Lu_Wang

What can I do to converts the DNA sequence into its reverse and make the repetitive sequences all forwardly?

If your sequences are in mixed orientations, you can do one of two things:

1. use the classify-consensus-vsearch method to classify your sequences, as this can classify reads in either orientation.
2. RESCRIPt has an orient-seqs method (see tutorial below) that will orient sequences relative to a reference database. This will allow you to orient your query sequences by comparing them to a reference database (which should presumably be all in forward direction), and then use classify-sklearn to classify those sequences.

Processing, filtering, and evaluating the SILVA database (and other reference sequence data) with RESCRIPt

Please consider this tutorial a living document, which may change based upon community feedback and ongoing plugin development. RESCRIPt RESCRIPt (REference Sequence annotation and CuRatIon Pipeline) is a python package and QIIME 2 plugin for formatting, managing, and manipulating sequence reference databases. This package was designed for compiling, manipulating, and evaluating sequence reference databases from SILVA, NCBI, Greengenes, GTDB, and other sources, and for constru…

Let us know which method you use and what you find. Good luck!