Binary measures as OTU values - appropriate for LEfSe?

[Seweryn Bialasiewicz]

Hello all,

 

I was wondering if anyone could comment on the appropriateness of using LEfSe on a binary "microbiome" dataset. That is, a dataset where each sample has a presence/absence score for ~50 organisms identified by bacterial culture and species-specific PCR rather than the standard OUT read counts/abundances. Essentially, each bacterial species would be treated as an OTU with a value of 0 or 1.

 

I'm just dipping my toes into this particular type of analyses, so it would be great to get counsel from people more experienced in the area.

 

Many thanks in advance,

-Seweryn