Search found 530 matches

by katharina
Fri Nov 20, 2015 1:14 pm
Forum: Predicting Genes with AUGUSTUS
Topic: AUGUSTUS for transcriptome
Replies: 15
Views: 9109

Re: AUGUSTUS for transcriptome

by suzuki on 18.09.2012 - 04:38
I trained with the Genbank file for Paramecium tetraurelia, available at
http://paramecium.cgm.cnrs-gif.fr/downl ... n_v1.gb.gz
Could you reproduce the same accuracy results?
by katharina
Fri Nov 20, 2015 1:14 pm
Forum: Predicting Genes with AUGUSTUS
Topic: AUGUSTUS for transcriptome
Replies: 15
Views: 9109

Re: AUGUSTUS for transcriptome

by Mario on 03.09.2012 - 15:28 As far as the segmentation fault is concerned, I believe I fixed it in the current working code. Thanks for sending the backtrace. In the next update it should not happen anymore. By 'Edit the Makefile so that the CFLAGS include -g -ggdb' I meant that the options -g -...
by katharina
Fri Nov 20, 2015 1:13 pm
Forum: Predicting Genes with AUGUSTUS
Topic: AUGUSTUS for transcriptome
Replies: 15
Views: 9109

Re: AUGUSTUS for transcriptome

by suzuki on 29.07.2012 - 10:05 According to the manual "Training AUGUSTUS" (http://bioinf.uni-greifswald.de/augustus/binaries/tutorial/training.html), I trained with Paramecium tetraurelia (ptetraurelia_annotation_v1.gb) on Linux. randomSplit.pl ptetraurelia_annotation_v1.gb 100 grep -c L...
by katharina
Fri Nov 20, 2015 1:13 pm
Forum: Predicting Genes with AUGUSTUS
Topic: AUGUSTUS for transcriptome
Replies: 15
Views: 9109

Re: AUGUSTUS for transcriptome

by suzuki on 28.07.2012 - 05:54 Because I didn't understand 'Edig the Makefile so that the CFLAGS include -g -ggdb', I didn't do it. Then, augustus worked on Linux, but augustus terminated prematurely and printed 'Segmentation fault: 11' on Mac OS 10.7.4. On Mac OS 10.7.4, running command in the de...
by katharina
Fri Nov 20, 2015 1:12 pm
Forum: Predicting Genes with AUGUSTUS
Topic: AUGUSTUS for transcriptome
Replies: 15
Views: 9109

Re: AUGUSTUS for transcriptome

by Mario on 27.07.2012 - 15:07 I am glad, it seemed to work out. Thanks for hanging in. I don't have a script for you that picks the longest/beest ORF. You have to train it yourself if you want species specific parameters for Paramecium tetraurelia. Please have a look at the tutorial and the other ...
by katharina
Fri Nov 20, 2015 1:12 pm
Forum: Predicting Genes with AUGUSTUS
Topic: AUGUSTUS for transcriptome
Replies: 15
Views: 9109

Re: AUGUSTUS for transcriptome

by suzuki on 27.07.2012 - 03:00 It worked when I compiled (cd src; make) after replacing igenicmodel.cc. augustus predicted 4048 CDS in 4011 of the 5230 mRNAs (Pte.seq.uniq). Of the 4048 CDS, 2723 contain start_codon, and 3773 contain stop_codon. Do you have any script to pick the longest ORF in ca...
by katharina
Fri Nov 20, 2015 1:12 pm
Forum: Predicting Genes with AUGUSTUS
Topic: AUGUSTUS for transcriptome
Replies: 15
Views: 9109

Re: AUGUSTUS for transcriptome

by Mario on 26.07.2012 - 18:54 Please double check. There should be one: wget http://bioinf.uni-greifswald.de/augustus/binaries/augustus.2.6.1.tar.gz wget http://bioinf.uni-greifswald.de/augustus/binaries/species/igenicmodel.cc tar xzf augustus.2.6.1.tar.gz diff src/igenicmodel.cc igenicmodel.cc 27...
by katharina
Fri Nov 20, 2015 1:12 pm
Forum: Predicting Genes with AUGUSTUS
Topic: AUGUSTUS for transcriptome
Replies: 15
Views: 9109

Re: AUGUSTUS for transcriptome

by suzuki on 26.07.2012 - 12:05 I am using augustus version 2.6.1 (http://bioinf.uni-greifswald.de/augustus/binaries/augustus.2.6.1.tar.gz). 'diff' command can't find any differences between two files: igenicmodel.cc of the current release 2.6.1 and this file (http://bioinf.uni-greifswald.de/august...
by katharina
Fri Nov 20, 2015 1:11 pm
Forum: Predicting Genes with AUGUSTUS
Topic: AUGUSTUS for transcriptome
Replies: 15
Views: 9109

Re: AUGUSTUS for transcriptome

by Mario on 26.07.2012 - 11:05 I had tried the commands from my last post above on your data and it worked. What version of augustus are you using? Have you replaced igenicmodel.cc? Can you please try this: Edig the Makefile so that the CFLAGS include -g -ggdb Please recompile then with make clean ...
by katharina
Fri Nov 20, 2015 1:11 pm
Forum: Predicting Genes with AUGUSTUS
Topic: AUGUSTUS for transcriptome
Replies: 15
Views: 9109

Re: AUGUSTUS for transcriptome

by suzuki on 25.07.2012 - 03:17 ~/augustus/src/igenicmodel.cc was replaced accordingly. cp ~/augustus/config/extrinsic/extrinsic.cfg . and edited the intron lines in extrinsic.cfg accordingly. touch hints.gff Then running augustus --species=tetrahymena Pte.seq.uniq --extrinsicCfgFile=extrinsic.cfg ...
by katharina
Fri Nov 20, 2015 1:10 pm
Forum: Predicting Genes with AUGUSTUS
Topic: AUGUSTUS for transcriptome
Replies: 15
Views: 9109

Re: AUGUSTUS for transcriptome

by Mario on 24.07.2012 - 18:35 You prevent the predictions of introns (while maintaining the ability to predict truncated CDS) like this: cp ~/augustus/trunks/config/extrinsic/extrinsic.cfg . touch hints.gff edit the intron lines in extrinsic.cfg so it looks like this intronpart 1 1e-100 M 1 1e+100...
by katharina
Fri Nov 20, 2015 1:10 pm
Forum: Predicting Genes with AUGUSTUS
Topic: AUGUSTUS for transcriptome
Replies: 15
Views: 9109

Re: AUGUSTUS for transcriptome

by Mario on 24.07.2012 - 16:57 I happened to just find an error that only happens in the case of --species=tetrahymena and should result in nonsense predictions, or let's rather say, undefined behavior. This only concerns tetrahymena as this uses different pattern sizes for the non-coding and intro...
by katharina
Fri Nov 20, 2015 1:10 pm
Forum: Predicting Genes with AUGUSTUS
Topic: AUGUSTUS for transcriptome
Replies: 15
Views: 9109

Re: AUGUSTUS for transcriptome

by suzuki on 24.07.2012 - 07:50 augustus --species=tetrahymena --genemodel=intronless Pte.seq.uniq > augustus.abinitio.gff getAnnoFasta.pl augustus.abinitio.gff augustus predicted 3588 CDS in 3574 of the 5230 mRNAs (Pte.seq.uniq), and thus I got more than 1 ORF in some of the mRNAs. Do you have any...
by katharina
Fri Nov 20, 2015 1:09 pm
Forum: Predicting Genes with AUGUSTUS
Topic: AUGUSTUS for transcriptome
Replies: 15
Views: 9109

Re: AUGUSTUS for transcriptome

by Mario on 23.07.2012 - 17:46 I haven't tried or evaluated augustus for predicting the ORF in an mRNA fragment, but here is my suggestion. augustus --species=tetrahymena --genemodel=intronless augustus deals with truncated genes by default, so that is not a problem. A CDS may be both left- and rig...
by katharina
Fri Nov 20, 2015 1:09 pm
Forum: Predicting Genes with AUGUSTUS
Topic: AUGUSTUS for transcriptome
Replies: 15
Views: 9109

AUGUSTUS for transcriptome

Originally posted in the old forum by suzuki on 22.07.2012 - 07:05 The ciliate Paramecium tetraurelia is closely related to Tetrahymena thermophilia (translation_table 6). The Genbank file for Paramecium tetraurelia is available at http://paramecium.cgm.cnrs-gif.fr/download/fasta/ptetraurelia_annot...
by katharina
Fri Nov 20, 2015 1:08 pm
Forum: Predicting Genes with AUGUSTUS
Topic: Branch sites in introns (and other conserved regions) used for gene prediction
Replies: 1
Views: 2184

Re: Branch sites in introns (and other conserved regions) used for gene prediction

by Mario on 21.05.2013 - 17:03 Yes, Augustus is using a branch point model and this is contributing a lot to accuracy. It is an inhomogeneous higher-order Markov chain model, that is trained using the fact that the distance to the 3' splice site is variable. This model is more general than the prof...
by katharina
Fri Nov 20, 2015 1:08 pm
Forum: Predicting Genes with AUGUSTUS
Topic: Branch sites in introns (and other conserved regions) used for gene prediction
Replies: 1
Views: 2184

Branch sites in introns (and other conserved regions) used for gene prediction

Originally posted in the old forum by EJ Blom on 25.03.2013 - 16:10 Dear developers, I was wondering whether conserved regions such as branch sites are used in the de novo gene prediction. I stumbled across this database:http://lemur.amu.edu.pl/share/ERISdb/weblogos_U2.html and there seem to be som...
by katharina
Fri Nov 20, 2015 1:07 pm
Forum: Predicting Genes with AUGUSTUS
Topic: augustus command for EVM
Replies: 1
Views: 2352

Re: augustus command for EVM

by katharina on 23.05.2014 - 15:56 I don't have much experience with EVM. But what matters is that you have the best possible gene set produced by AUGUSTUS, i.e. you should use parameters that suit well for your species you should use hints from extrinsic evidence if available generally: make sure ...
by katharina
Fri Nov 20, 2015 1:07 pm
Forum: Predicting Genes with AUGUSTUS
Topic: augustus command for EVM
Replies: 1
Views: 2352

augustus command for EVM

Originally posted in the old forum by megha on 22.05.2014 - 11:43
What is the ideal input command for augustus which can be put as input for EVM.
by katharina
Fri Nov 20, 2015 1:07 pm
Forum: Predicting Genes with AUGUSTUS
Topic: hints priority
Replies: 2
Views: 2476

Re: hints priority

by katharina on 21.06.2014 - 15:37 It depends on what you want to achieve. If you think that ESTs are more reliable than protein hints, then you should assign a higher priority to the ESTs, yes (that means a higher number). The hint priority determines which hint will be used in case of a contradic...