Any pipeline to find automatically ORF in consensus sequences?

[Christopher Sauvage]

Dear members,

Following my 454 data assembly, sequenced from a cDNA library, I'm looking for an "automated way" to find ORF for all the consensus contig sequences. My assembly was performed with a high percentage of similarity (0.98) using CLC workbench. Unfortunately, in my favorite species, there is no complete genome available at this moment to gather any information.

So, did anyone have already experienced this problem? How to find the true ORF with 454 contig that may contains some errors, such as false-positive insertion that shift the frame?

Thanks a lot,
Regards,
Christopher

[steven]

I think that FrameD and FrameDP can provide you exactly what you are looking for: it is a gene finding software that is designed to be robust regarding sequencing errors. If you are not working on prokaryotic genome, look for its "big brother" EuGene..
have fun,
s.

[Christopher Sauvage]

Hi Steven

Thanks for your help! It is really appreciated. On my side, I have found that ORF predictor may perform a similar job.
Visit: http://proteomics.ysu.edu/tools/OrfPredictor.html for more details.

I'll let users know how things are going on with these different softwares.

Best regards,
Christopher

[kmcarr]

You might also try ESTScan:

Software: http://estscan.sourceforge.net/
Paper: http://citeseerx.ist.psu.edu/viewdoc...=rep1&type=pdf

[steven]

Christopher, are you sure that ORF predictor is resistant to sequencing errors and partial ORF?
Thanks kmcarr, i didn't know ESTScan.
I think that FrameDP can train itself (discriminative learning), deal with noisy sequences, partial ORFs, and high variations in GC content. Plus it is not restricted to pure "ab initio" prediction, as it can integrate protein similarities. They have a web server, if you want to give it a try..
good luck,
s.

[JacquesWvdH]

Hi all, I am new in this field of research.
I have contigs from acinomyces and I want to detect and annotate ORFs to be viewed in artemis.
I am focussing on acinomyces (and eucaryotes in the future) and I am considering the use of FramdeDP with a local NCBI-database for blasts.
Would you agree with this choice or should I consider other options?

[Christopher Sauvage]

Hi Jacques,

I have tested the FrameDP and ORFPredictor softwares to find ORF in my NGS data. ORF predictor has reached my expectation as it is based on the blastx results of your consensus sequences to predict your (partial) ORF. It can run your analysis in batches. Moreover, when there are sequencing or assembling errors, ORF predictor is able to predict an ORF with a frame shift.

Regards.