When I was using the knownCanonical table to find the canonical transcript
for MYC, I find that there are two entries.  See below.  I also included
some fields from hg19.kgXref fields.  Is there an accepted method to
determine which one is the most "canonical" transcript?  Perhaps use the
transcript where there is a "NM" as the prefix in refseq?

Thanks,
Andrew

#hg19.knownCanonical.chrom hg19.knownCanonical.chromStart
hg19.knownCanonical.chromEnd hg19.knownCanonical.clusterId
hg19.knownCanonical.transcript hg19.knownCanonical.protein hg19.kgXref.kgID
hg19.kgXref.mRNA hg19.kgXref.spID hg19.kgXref.spDisplayID
hg19.kgXref.geneSymbol hg19.kgXref.refseq

chr8    128748314       128753678       24861   uc003ysi.2      uc003ysi.2      
uc003ysi.2      NM_002467       A0N2G3  A0N2G3_HUMAN    MYC     NM_002467
chr8    128806778       129113498       24862   uc010mdq.2      uc010mdq.2      
uc010mdq.2      NR_003367                       MYC     NR_003367
_______________________________________________
Genome maillist  -  [email protected]
https://lists.soe.ucsc.edu/mailman/listinfo/genome

Reply via email to