[Genome] refseq table

[Mike Mitchell]

Hello Jean-Baptiste,

The multiple entries per gene are as a result of the existence of known
splice variants for genes, each of these will have it's own RefSeq entry. A
gene symbol does not necessarily uniquely identify a unique cDNA/mRNA
sequence.

In your example, GAS2L1, there are three known splice variants represented
by the RefSeq IDs NM_006478, NM_152236 and NM_152237. Thus there are a
different number of exons, and a difference in some of the exon lengths.

I hope that this helps.

On 9/5/07 15:26, "Jean-Baptiste" <veyrieras at moulon.inra.fr> wrote:

> Dear,
> 
> I am quite a newbie to USCS browser. I am particularly interested by
> downloading the information about human gene structure (exon, intron,
> promoter region, etc...).
> I have downloaded the refFlat.txt file and I don't understand why some
> genes have several entries (i.e different number of exons and eventually
> different exon positions). For ex :
> 
> GAS2L1  +       28027550        28033328        28028649
> 28033040        6
> 28027550,28028630,28030982,28031181,28031370,28032005,
> 28027730,28029282,28031090,28031278,28031542,28033328,
> GAS2L1  +       28027550        28033328        28028649
> 28031546        5       28027550,28028630,2803
> 0982,28031181,28031370,     28027697,28029282,28031090,28031278,28033

-- 
Mike Mitchell
Bioinformatics & Biostatistics Service
Cancer Research UK +44 (0) 207 269 3115