AUGUSTUS Forum

Originally posted in the old forum by Julia on 09.01.2014 - 15:34
Hello,
I was wondering if I could use AUGUSTUS to predict genes from large contigs obtained from a metagenome assembly. I mean, I have some eukaryotic contigs (from different unknown species) and I want to train AUGUSTUS with genes within these contigs obtained by CEGMA (.gff file) and run it to find new genes. Do you think it will work despite contigs and the training set are a mix of different genomes?

by Katharina on 16.07.2014 - 14:51
So far, we do not have a "generalized" (i.e. species-unspecific) parameter set for AUGUSTUS. I have no idea how well it will work. The more species are in the contigs, the less closely related they are, the worse it will work, I assume.
Katharina