Estimation of prediction accuracy

Discussions about predicting genes with AUGUSTUS. Not covered here: WebAUGUSTUS and BRAKER1

Moderator: bioinf

Post Reply
User avatar
katharina
Site Admin
Posts: 531
Joined: Wed Nov 18, 2015 6:14 pm
Location: Greifswald
Contact:

Estimation of prediction accuracy

Post by katharina »

Originally posted in the old forum by Sammy on 01.07.2013 - 22:08
How can we estimate the accuracy of gene prediction if we are using species A parameters for prediction of species B?
User avatar
katharina
Site Admin
Posts: 531
Joined: Wed Nov 18, 2015 6:14 pm
Location: Greifswald
Contact:

Re: Estimation of prediction accuracy

Post by katharina »

by katharina on 09.07.2013 - 15:23
First of all, in order to measure gene prediction accuracy in your target species, you need a reliable set of genes of your target species. You can construct such a gene set e.g. by using ESTs2genome or protein2genome alignments.
Then, you predicted genes in the reliable gene set using the parameter set that you like. Count how many nucleotides/exons/genes were predicted correctly as coding (TP), how many were wrongly predicted as coding (FP) and how many were wrongly not predicted as coding, i.e. overseen (FN).
Katharina
User avatar
katharina
Site Admin
Posts: 531
Joined: Wed Nov 18, 2015 6:14 pm
Location: Greifswald
Contact:

Re: Estimation of prediction accuracy

Post by katharina »

by Sammy on 10.07.2013 - 01:34
Thanks, are you telling about http://emboss.bioinformatics.nl/cgi-bin ... est2genome program?
User avatar
katharina
Site Admin
Posts: 531
Joined: Wed Nov 18, 2015 6:14 pm
Location: Greifswald
Contact:

Re: Estimation of prediction accuracy

Post by katharina »

by katharina on 10.07.2013 - 14:11
I have no experience with the EMBOSS est2genome programe. We usually use PASA. But est2genome might work as well. The main point is that you need to obtain full-length gene structures in gtf format - using whatever tool.
Katharina
Post Reply