[maker-devel] control file for SNAP training

[Marc Tollis]

This is a question about process, and to make sure I am doing things right
(when time is of the essence, some mistakes can set you back weeks).

I have run maker on my de novo vertebrate genome, using only the predictive
proteome from a congener (well-studied and available on Ensembl), and
generated the HMM for the first round of SNAP training. As per the 2014
tutorial, I edited the control file for this step as follows: I added the
path to the .hmm file, and set protein2genome to 0.

When I run maker, I notice that in addition to snap, it is still running
blastx and exonerate however. I noticed that this is because I did not
remove (or "comment out") the path to the protein.fa in the control file
(the output looks markedly different when I do comment out the protein file
- and I can't even tell if it's running snap in this instance).

Is it simply using exonerate to place the ab initio predictions on the
scaffolds (meaning that having protein2genome=1 is to tell maker to make
evidence annotations) ? Did I do this correctly, or should I also remove
the protein.fa out of the control file for SNAP training?
​
-- 
*Marc Tollis, Ph.D.*
*Post-Doctoral Research Associate*
*Arizona State University*
*LSE 313*
*(480) 965-7456*
marc.tollis at asu.edu

*website: *https://sites.google.com/site/tollisresearch/
*blog: *anolistollis.wordpress.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://yandell-lab.org/pipermail/maker-devel_yandell-lab.org/attachments/20150315/ef226da4/attachment-0002.html>