I'm trying to go through the supplementary material of Shah's Nature paper 2012 on triple negative breast cancer genomics. Apart from the tricky annotations (hg18 and discontinued gene symbols), I don't find a lot of features that have names like "LOC729608, LOC340947, LOC729317" etc. Where can I find what these regions are in hg19?
If I understand your question correctly, the issue is you cannot find the hg19 coordinates for the gene names you listed? For a short list, you can try searching for these particular annotations at www.ncbi.nlm.nih.gov/gene/
To approach it more systematically, I suggest you get the GeneCards annotation track from UCSC. http://genome.ucsc.edu/cgi-bin/hgTracks?db=hg19&hgt.customText=http://www.genecards.org/all_genecards.txt&position=chr10:38358984-38379805
Follow the link, and use the Table Browser to download the lookup table. It should have most of these "LOCxxxx" gene names with their corresponding genomic loci.
Daniel: that's exactly the issue. I tried all of these resources, but, just to give you a glimpse, the result is invariably "Sorry, couldn't locate LOC729608 in genome database" [source UCSC]. The point is: these nomenclatures are not in GeneCards, nor in UCSC or in Vega or in RefSeq or anywhere else to be found. Just try and google them and see what happens...
Daniel: PS for the sake of curiosity, the table I'm referring to is S3 of the Nature. 2012 Apr 4;486(7403):395-9 paper...
Hi Zoppoli. I've checked these particular gene symbols, and the Gene database in NCBI finds them. But as I indicated, it is easier to use the GeneCards reference table for a longer list of genes. Unfortunately, the UCSC browser does not support text searching for these annotations. However, if you follow the link I sent you above, go to Table Browser, Custom Tracks, GeneCards Track, you can download a text file with all the gene symbols in GeneCards mapped to hg19 coordinates. If you can't get the table to download, please let me know and I'll post it on GenomeSpace for you.
Thanks for specifying the table, Zoppoli. Since the exact hg18 coordinates for the mutations are provided, you can simple generate a bed file of all these sites, and use the Lift-Over tool to map to hg19. Simpler than trying to map gene annotations across assemblies.
Daniel: last question, after liftover, how do I reannotate with hg19 GeneSymbols?
Daniel: I downloaded the custom track GeneCards BED file as you suggested, but the LOCxxx genes are not there...
Zoppoli- the GeneCards BED file contains at least the following entries:
chr10 38368984 38369805 LOC340947 0 -
chr2 65432109 65433412 LOC729317 0 +
LOC729608 is not included, but the NCBI Gene database matches it to RPF2P1
chr20 34340717 34341659 RPF2P1 0 +
With respect to annotation of LiftOver files, you can simply include the original gene symbol from the Shah paper in your input as a 4th column. The output from the UCSC LiftOver conversion to hg19 will keep the gene symbol associated with its corresponding chromosomal locus.
One other note, Zoppoli, with respect to annotations. Hypothetical/predicted gene loci that are not found in the GeneCards BED file are expected to occur, because some candidate genes get revoked from the database (e.g. http://www.ncbi.nlm.nih.gov/gene/?term=LOC100128709) and others get renamed (e.g. http://www.ncbi.nlm.nih.gov/gene/?term=LOC100129128). Those that have been renamed can still be found by UCSC if you choose hg18 as your working genome. Hope this helps!
- Badges
- Science topic
- Similar topics
- Biological Science
- Genetics
- Genetic Analysis