Missing taxa

[ntromas]

Hi,

I have used Vamb the last years and I really like the pipeline. For the first time, despite having a pretty good depth (~50M/samples), I am not able to generate MAGs from a specific genus that I know is present from microscopy and from reads based approach like Sylph.
Contigs were generated with megahit and I used two binning methods (metabat2 and vamb Version: 5.0.4) + Das_Tool.
It is not a problem from Vamb but just curious about potential element that would make impossible the recovery of MAGs from this genus (high micro-diversity?).

Cheers,

Nico

[simonrasmu]

Hi Nico,

This could be a use case for our annotation enhanced binner TaxVAMB which use contig taxonomic annotations to improve binning. If your some representatives of the genus of interest is in the database you should expect to see the improvements

Cheers

Simon

[ntromas]

Thanks Simon for the quick answer. So I need to taxonomically classified my contigs and used the output of that with TaxVamb. Any tools recommendations for that? I usually use CAT but it will be very long with such dataset.

Cheers,

Nico

[simonrasmu]

Yes, we use mmseqs taxonomy - maybe @jakobnissen or @sgalkina can help out with commands and/or databases