I have two genes around 12kbp apart in the xenopus genome. I want to check how distant the homologs of these genes are in different vertebrate taxa. Should I do a BLAT of the genomic sequence corresponding to both these genes (against each taxa I am interested in) and then choose the hits that have maximum sequence similarity ? I am pretty new to the UCSC genome browser so I guess there might be some better way of doing this that I dont already know of. The names for the genes are different in different taxa and hence, I was talking of doing BLAT to find the corresponding homolog. For eg, I have RHO and IFT122 in Xenopus but I dont get any of these genes in the Zebrafish gentrack.
Seems to me this would be easier to do without the UCSC genome browser if you plan on checking tens or hundreds of taxa. Use your two known translated genes as separate, high stringency BLASTX queries with
nr as the reference database. Pull out the best hits for each species and have a script compare their locations.
If you are only interested in a small number of taxa, and all of which are found in USCS, that would be fine too. In any case, you are right that searching gene names is not guaranteed to work, and that sequence homology is usually the way to go.
Using your example genes and BLATing the RefSeq predicted amino acids against Mouse in the UCSC browser,
RHO is found on Mouse
chr6 115932004-115935635 and
IFT122 is found on Mouse
chr6 115862712-115926307. This gives a distance between the two genes of only
5,697bp. To prove the legitamacy of this method, these coordinates in Mouse are actually already annotated as Rho and IFT122 respectively, so it surely works.