Friday, October 30, 2009

New release is here

The October 2009 release of OMA with 907 genomes is now online. The interruption of the OMA Browser was only for a short time, now everything should work well.

There is a slight change in the way alternative splicings are treated. From the current release on, only one splicing variants of a gene can have orthologs (we chose the one with the most matches in other genomes). This should have the effect that groups are more complete> Before, the different splicings variants had the tendency to be scattered among several groups.

And finally, we would like to point out again the survey on the OMA Browser. We really like to know your opinion so we can improve the OMA Browser. And you have the chance to win an iPod. Go here to fill out the survey, it won't take more than 5-10 minutes.
Posted by at

2 comments:

  1. vacalactica-roblesNovember 3, 2009 at 2:29 PM

    Hi Oma guys, thank you for the update!!! I have a couple of questions for you:
    1) why did you use MAFFT to align the aa sequences? did you try other alignment programs and finally you choose that one?
    2) It is possible to download the multiple sequence alignment for all oma groups?
    3) if you translate the cDNA sequences you have available, the product is exactly the same you have in the protein sequence files?

    thank you in advance,

    all the best

    juan

    ReplyDelete
  2. Adrian SNovember 12, 2009 at 2:12 PM

    Thank you for the questions. Here are the answers (I emailed them also to you):

    1) The MSAs are just for visualization of the group. The OMA algorithm works on pairwise alignment. (To be precise, "local" alignments where only the best matching parts of the sequences are aligned, whereas the MSA display the alignments of the whole sequences.) We use Mafft since it is quite fast, as we don't want to spend to much CPU time on the displaying of the MSA.

    2) Unfortunately no. The MSAs are computed only the first time somebody would like to see them. So most of them don't even exist (yet). If it is really helpful for your project, however, we can certainly help you to compute the MSAs. But it is a lot of CPU time that would be used.

    3) Yes, we made sure that the cDNA and the proteins match in all cases. Sometimes there are Xs in the cDNA which of course cannot be translated to an amino acid. Furthermore, keep in mind that there are
    different genetic codes.

    ReplyDelete

Subscribe to: Post Comments (Atom)