Question: Exonerate: Issues please help
gravatar for carlosalfonsogonzalez6
14 months ago by

Hello Everyone,

Im currently facing some problems with exonerate(and tried version 2.2 and 2.4), using the next code:

exonerate --model est2genome --softmasktarget yes --bestn 1 --minintron 20 --maxintron 20000 --bigseq yes --query Long_isoform_CDNA.fasta --showvulgar no --target DvirgenomeMasked.fa.out --showalignment no --showtargetgff yes --verbose 0 > AnnoDvir_clean_percent.gff

Persons who have used this program, How do you overcome the slowness problem? I have also see that when i do not used the "--bestn" command wich gives the best aligned secuence the code runs perfeclty.

Hope you can help me with that Thanks Greetings

rna-seq alignment • 688 views
ADD COMMENTlink modified 13 months ago by Michael Dondrup45k • written 14 months ago by carlosalfonsogonzalez60

How did you install the versions of Exonerate you used?

ADD REPLYlink modified 14 months ago • written 14 months ago by h.mon23k

Following the user guide, ./configure, make, make installl, Do you have an idea about making the programing runnning faster?

ADD REPLYlink written 13 months ago by carlosalfonsogonzalez60

Hi I havent find an answer yet

ADD REPLYlink written 13 months ago by carlosalfonsogonzalez60

Did you check that you're not running out of memory ? --bestn is supposed to speed up exonerate. However, you're using quite a high value for it which may make exonerate work more than needed. Since --model est2genome uses essentially a Smith-Waterman alignment algorithm, --bestn 70 probably forces exonerate to run this for quite a lot of sequences. If the problem is due to the --bestn option then you may be better off not using it and then post processing the output. Also look at the section on 'Strategies for speed' in the man pages.

ADD REPLYlink written 13 months ago by Jean-Karim Heriche18k

Oh, Thank you very much for you answer i use --bestn 1 i dont know what i put it that way on my question, Does you answer still the same? Im trying to map cDNA from Drosophila virilis to the Drosophila melanogaster genome, this with the goal to obtain an anotation for both species... for rna seq analyses. Im open to suggestions.


ADD REPLYlink written 13 months ago by carlosalfonsogonzalez60
gravatar for Michael Dondrup
13 months ago by
Bergen, Norway
Michael Dondrup45k wrote:

The debian man-pagel also lists a --cores parameter. You can also try setting the --score parameter and setting --fsmmemory to be close to physical RAM.

Using GMAP could be an alternative.

From the manual:

STRATEGIES FOR SPEED Keep all data on local disks.

Apply the highest acceptable score thresholds using a combination of --score, --percent and --bestn.

Repeat mask and dust the genomic (target) sequence. (Softmask these sequences and use --softmasktarget).

Increase the --fsmmemory option to allow more query multiplexing.

Increase the value for --seedrepeat

When using an alignment model containing introns, set --geneseed as high as possible.

If you are compiling exonerate yourself, see the README file supplied with the source code for details of compile-time optimisations.

ADD COMMENTlink modified 13 months ago • written 13 months ago by Michael Dondrup45k
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 2340 users visited in the last hour