RNASeq data for training and hintsfile

Discussions about BRAKER, the automated joint usage of GeneMark and AUGUSTUS.

Moderator: bioinf

Post Reply
User avatar
katharina
Site Admin
Posts: 530
Joined: Wed Nov 18, 2015 6:14 pm
Location: Greifswald
Contact:

RNASeq data for training and hintsfile

Post by katharina »

Originally posted in the old forum by chrikus on 20.03.2015 - 09:18
hey,
I have several sets of RNAseq data and map it with segemehl.
I´m trying to follow the manuals of incorparating RNAseq, but I don´t know, how to solve the problem with the several data sets.
1) Can I use the merged sets as input for augustus?
or is it better to merge it afterwards?
2) How can I prepare RNAseq data for training?
different approach:
is it possible to create hintsfile with PASA output? or PASA can only be used for training?
thanks in advanced,
chrikus
User avatar
katharina
Site Admin
Posts: 530
Joined: Wed Nov 18, 2015 6:14 pm
Location: Greifswald
Contact:

Re: RNASeq data for training and hintsfile

Post by katharina »

by katharina on 20.03.2015 - 12:32
Last time I checked, the bam format for "segemehl" had some properties that are not in the standard bam format. I am not sure whether our tools (which only support standard bam format) are compatible with the segemehl format, today.
1) Yes, use the merged bam files for creating hints.
2) I recommend you have a look at BRAKER: http://bioinf.uni-greifswald.de/augustu ... /index.php (BRAKER requires the same standard bam format as all our tools).
It is possible to create a hints file from PASA. (You'll need to script that, yourself, but it should be easy.)
Katharina
Post Reply