Hello, You are correct - if the hg19 UCSC Genes track was recreated today, then the HUGO gene symbol would have been used. However, it was not available at the time of the track build.
If you search for LOC401629 in an earlier version of the human assembly at UCSC (hg18), the RefSeq record reflects the intact prior version of the RefSeq and the assigned gene symbol of LOC401629. UCSC Genes track updated in hg19: Data last updated: 2009-10-08 RefSeq last updated (all assemblies): PRI 15-FEB-2010 Link into RefSeq (current): http://www.ncbi.nlm.nih.gov/nuccore/NR_002160.1 As a work-around to find the UCSC Gene in hg19 with the new RefSeq identifier, do the following: 1) open the gateway page for hg19 2) enter "NCRNA00230A" and submit 3) a choice of three links is given, use either of the two under "RefSeq Genes" (this gene has a known duplication on chrY) 4) once in the browser, open the UCSC Genes track to find the associated UCSC Gene based on the provisional version of the RefSeq. We are very sorry for the inconvenience. We are not able to update UCSC Genes track at the same rate that RefSeq updated due to the complexity of the processing and the dependency on multiple datasets. However, the RefSeq Genes data is taken directly from GenBank as a nightly automated process. Using the two together will help you access the most current and complete annotation. Hopefully this helps, Jennifer --------------------------------- Jennifer Jackson UCSC Genome Informatics Group http://genome.ucsc.edu/ On 4/13/10 9:19 AM, Bio X2Y wrote: > Hi, > > I understand that the UCSC Genes annotation process consults a range of > sources to assign a "genesymbol" to each transcript. > > In an earlier post ( > https://lists.soe.ucsc.edu/pipermail/genome/2006-April/010350.html), Fan > outlined the general strategy - use the RefSeq symbol if a transcript's > representative mRNA is a RefSeq, otherwise consult UniProt, etc. Finally, if > no symbol can be assigned, use the ID of the mRNA. > > I'm looking at "uc004ftj.2" (hg19), which has a gene symbol of 'LOC401629'. > > This transcript seems to be based on a RefSeq (the 'refseq' column in kgxref > is NR_002161, and the exons structure matches very closely). That RefSeq > record has a gene name of 'NCRNA00230A', which is a HUGO symbol. > > I'd appreciate it someone might be able to explain the reason why in this > case, the RefSeq symbol was not used, and perhaps where LOC401629 was > derived from? > > Thanks for your time. > _______________________________________________ > Genome maillist - [email protected] > https://lists.soe.ucsc.edu/mailman/listinfo/genome _______________________________________________ Genome maillist - [email protected] https://lists.soe.ucsc.edu/mailman/listinfo/genome
