Hello Halian,
abParts was made in the following manner:
All genbank records were searched with the following code to detect
those that had "immunoglobulin" and "chain"
in the description and one of quite a number of other terms to make the
search more specific.
if (stringIn("immunoglobulin", description) && stringIn("chain",
description))
> {
> if (stringIn("variable", description) || stringIn("vdj", description)
> ||
> stringIn("vlj", description) || stringIn("vhdj", description) ||
> stringIn("v region", description) || stringIn("vj region",
> description) ||
> stringIn("v-region", description) || stringIn("v-j-c region",
> description) ||
> stringIn("v-d-j", description) || stringIn("rearrange",
> description) ||
> stringIn("v-d-t region", description) || stringIn("v-j region",
> description) ||
> stringIn("v-d-c region", description) || stringIn("kappa chain
> (igk) mrna, vk", description) ||
> stringIn("partial mrna for immunoglobulin", description) ||
> stringIn("nonfunctional immunoglobulin", description) ||
> stringIn("kappa light chain (igvk-a", description) ||
> stringIn("heavy chain complementarity-determining region 3",
> description) ||
> stringIn("chain mrna, partial cds", description) ||
> endsWith(description, "immunoglobulin mu heavy chain-like mrna,
> partial sequence.") ||
> endsWith(description, "immunoglobulin mu heavy chain mrna,
> partial sequence.") ||
> endsWith(description, "nonfunctional immunoglobulin heavy chain
> mrna, partial sequence.") ||
> endsWith(description, "fab mrna, partial cds.") ||
> endsWith(description, "immunoglobulin e heavy chain (ige) mrna,
> partial cds."))
The resulting records were identified in the full genbank vs. reference
human genome alignment, and the alignments clustered.
The four largest clusters, which cover the three well identified
antibody variable regions, were kept. (One of the ab regions clusters
into two parts).
These are in the UCSC Genes track because, they are genes, albeit
strange ones. And they are combined into one record because it seemed
like it would clutter up things too much if we had a separate gene for
each fragment.
Best regards,
Pauline Fujita
UCSC Genome Bioinformatics Group
http://genome.ucsc.edu
On 03/09/11 08:51, Halian Vilela wrote:
> Hey folks, my first post here, I would like to know if anyone has any
> further informations about the "abParts" gene sequence (gene ids:
> uc010tyt.1, uc002stl.2, uc010fhm.2, uc010ytr.1
> and uc011aim.1 ).
>
> I've tried to get more information about this on the description page, but
> that's too strange, all the links to external websites like ENSEMBL or H-INV
> leads to different genes names and ID's, it makes no sense at all.
>
> I understand that it should depict parts of antibodies, but how this was
> assembled in one huge sequence ? And why it is shown as a reference like it
> was a real gene ?
>
> Please, any information would be greatly appreciated.
>
> Thanks,
> Halian
> _______________________________________________
> Genome maillist - [email protected]
> https://lists.soe.ucsc.edu/mailman/listinfo/genome
_______________________________________________
Genome maillist - [email protected]
https://lists.soe.ucsc.edu/mailman/listinfo/genome
