Hello Peng, You are correct, some of the other transcripts in the UCSC Genes track do map to more than one location. Why both are retained is explained in the track description. Which mapping to use will be up to the scientist using the data to decide.
Thanks! Jen --------------------------------- Jennifer Jackson UCSC Genome Informatics Group http://genome.ucsc.edu/ On 4/28/10 8:26 PM, Peng Yu wrote: > On Wed, Apr 28, 2010 at 7:31 PM, Jennifer Jackson<[email protected]> wrote: >> Hi Peng, >> >> The UCSC Genes track is more discriminatory, with only one mapping. Using >> the UCSC mapping is probably the best way to go. >> >> database=mm9, search position box = NR_002891 >> >> UCSC Genes >> >> NR_002891 (uc008guv.1) at chr19:12979721-12981475 - NR_002891 > > Although UCSC genes give a unique location. But this is not always > true. UCSC gene corresponds to NM_001025388 also maps to multiple > locations with different exons in each location. > > I download the exon table from refFlat. The following refseqs map to > more than one genomic locations and the number of exons are not the > same across the multiple genomic locations. > > [1] "NM_001025388" "NM_001099313" "NM_001101644" "NM_001166638" "NM_001166639" > [6] "NM_001166641" "NM_001170345" "NM_001177538" "NM_001177539" > "NM_001177543" > [11] "NM_009286" "NM_027087" "NR_002445" "NR_002883" "NR_002889" > [16] "NR_002891" "NR_003596" > > In general, can I assume these multiple mapped transcripts with very > different intron/exon structure unreliable? If they map to multiple > genomic locations but the intron/exon structures are the same at these > locations, these mapping should be considered reliable? > >> RefSeq Genes (refGene) >> >> NR_002891 at chr19:12979721-12981475 >> NR_002891 at chr10:5917375-5943370 >> >> (The RefSeq hits are close in score, which is why both were retained. Please >> note that the Methods for the track may have a mistake - we'll get back to >> you about that). >> >> Thanks, >> Jen >> >> --------------------------------- >> Jennifer Jackson >> UCSC Genome Informatics Group >> http://genome.ucsc.edu/ >> >> On 4/28/10 2:07 PM, Peng Yu wrote: >>> >>> Hi, >>> >>> According to ucsc genome browser, NR_002891 maps to the following >>> genomic location. These two mappings show very different number exons. >>> One is 1, the other is 12. >>> >>> I don't understand why a single transcript could come from two genomic >>> location that have very different intro/exon structures. Are these >>> mappings real? Or one of them is an artifact. Or the underlining >>> genome assembly (mm9) is defective at one of these locations? >>> >>> NR_002891 at chr19:12979721-12981475 >>> NR_002891 at chr10:5917375-5943370 >>> >>> -- >>> Regards, >>> Peng >>> _______________________________________________ >>> Genome maillist - [email protected] >>> https://lists.soe.ucsc.edu/mailman/listinfo/genome >> > > > _______________________________________________ Genome maillist - [email protected] https://lists.soe.ucsc.edu/mailman/listinfo/genome
