Gene Prediction: AUGUSTUS

[Amanda_G]

Hello All,

I have sequenced and assembled the chloroplast genome of a red alga. I would like to use AUGUSTUS for gene prediction within my genome; however, I would like to choose an organism model as close to my species as possible. Does anyone know if there's a way to specify an organism model that's not already listed in AUGUSTUS? The model organisms I would like to use have fully assembled and annotated genomes.

Thanks!

- Amanda G

[WhatsOEver]

If it is not yet available in AUGUSTUS, you may submit a new training request online: http://bioinf.uni-greifswald.de/weba...raining/create

[jordana]

I'm trying to train and predict UTR in transcriptome using Augustus.
I'm following the script autoAug.pl (in augustus.2.5.5 version).

http://augustus.gobics.de/binaries/scripts/README.autoAug

My command line:
sudo AUGUSTUS_CONFIG_PATH=home/augustus.2.5.5/config perl
/home/augustus.2.5.5/scripts/autoAug.pl --species=myspecie
--genome=genome.fa --trainingset=merged_annotations.gff --cdna=transc.fasta
-v --noninteractive --workingdir=/home/jordana/teste25 > GFF_step1

But, in first step (-- Construct an initial training set of genes (protein
coding regions thereof), e.g. using PASA) there are a something that stoped
the script. And every attempt the log (GFF_step1 file) returns and stop in
this part "2 First try with etraining: etraining --species=myspecie
training.gb.train >train.out ...". Furthermore, below command line appear
just "failed to execute:".

I also tried use another data set (e.g. human genome and genbank human
genome), but I had the same problem. And I not have ideia what happens.

[AB.PATEL]

Hello,

I want to predict gene in buffalo dataset, which training dataset should I use? Do I need to train cattle data or human training dataset will work?

Thanks