Eukaryotic orf finder

[svj]

Hi All,

I am looking for Eukaryotic orf finder algorithm/source code. I am trying to build training model for unknown eukaryotic genome using Glimmerhmm. I need collect orf's for the Glimmerhmm training model. So I did BLASTp against known eukaryotic protein sequences (closest neighbour to the unknown eukaryote) but am unable to build the training model with resultant orf's. The error I get after trainGlimmerhmm is:
Training data created successfully! Check exons.dat and seqs for accuracy.


Acceptor sites for training: 18292
False acceptor sites for training: 853751
Donor sites for training: 18219
False donor sites for training: 672464


ERROR 69: /GlimmerHMM/train/score exited funny: 35584


If this process of building training model is right then can anyone help me with this situation. If not then what can I do to build training model? Should I look for acceptor and donor sites in the upstream and downstream of the orf's I got in blastp?

[Torst]

This is a good starting point: http://en.wikipedia.org/wiki/Gene_prediction

Then look at the software in the References section. Augustus might be nicer.

[vikas0633]

I have same error. Let me know if you find a solution for glimmer

[ersgupta]

anybody has a solution for this error?

[colindaven]

Augustus is easier people, I would recommend it over Glimmer.

[fahmida]

I also agree with Torsten and Colin. Augustus is much easier to deal with.