Hi, I am screening a genome collection for a protein of interest. Previously, I have been firstly predicted CDSs with prodigal and used that as input. Now I thought about trying to avoid that step to streamline a bit, since I am using a larger dataset.

So the question is, if there is any reason I shouldn't blastx directly against the genomes? Outside of having more spurious matches, I can't think of drawbacks... On the other hand, is there any specific option I should use in the call? I thought of using --range-culling to avoid getting overlapping hits (if I understood the documentation properly).