[Genome] Understanding alternative splice data tables

[Clancy, Kevin]

Dear Sirs
I am interested in using Tables to extract information on alternative
spice exons in human chromosome 21. So I would like to have a measure of
for each exon seen by aligning ESTs along the genome, how many times
that exon is present in all the represented genomes and how many times
it binds to the upstream and downstream exons on either side of it.
Ideally I would like to use the information from the table to extract
all these alternative splice products from the genomic sequence with
some statistics on the prevalence of each exon triplet. Ultimately I'm
interested in generating a probability based tool to generate
alternative spice products.

Looking at the Spliced ESTs/intronEST table, I can see that you have
nucleotide sizes, start positions and block sizes  in both the query EST
and the chromosome. Should the sum of the block sizes add up to the
query EST size? If not, why is there a difference in the two? 

Secondly I have looked at the Sib Alt-Slicing/sib TX graph table and you
have a network representation of vertices and edges fields corresponding
to the EST but I don't quite understand how to use the information you
have there to tackle my problem. Any help or simple example you can
point me towards would be very appreciated.

Finally are these the best tables for me to look at to try and generate
this type of information? If not, where would be a better table?
Thanks kevin

Kevin Clancy, PhD
Senior Scientist, Informatic Sciences
Invitrogen Corp
Carlsbad, CA 92008
Phone: (240) 379 4401 x84401
Cell: (240) 417 8604
Email: kevin.clancy at invitrogen.com