two different UCSC isoforms for the same transcript

Question

[donegal77]

Hi,

a collaborator of mine is doing correlation analysis of gene expression using a RNA-seq data from TGCA but our protein of interest (the antiapoptotic Bcl-xL) derived from an alternative splicing event. Since we should distinguish out right isoform from the shortest one (Bcl-xS), which has the opposite cellular function, my collaborator downloaded the RNA-seq data specific for the different isoforms and we found that two different UCSC transcripts (the "main" uc002wwl.3 and uc002wwn.3) encodes for Bcl-xL. My question is related to the correct way to manage these data: should we add the contribution of both UCSC transcript or working with the mean of the two values for each sample? Thanks in advance,

Chiara

[Pierre Lindenbaum]

i've checked this:

$ mysql --user=genome --host=genome-mysql.cse.ucsc.edu -A -D hg19  -e 'select * from knownGene where name in("uc002wwl.3","uc002wwn.3") '
+------------+-------+--------+----------+----------+----------+----------+-----------+-----------------------------+-----------------------------+-----------+------------+
| name       | chrom | strand | txStart  | txEnd    | cdsStart | cdsEnd   | exonCount | exonStarts                  | exonEnds                    | proteinID | alignID    |
+------------+-------+--------+----------+----------+----------+----------+-----------+-----------------------------+-----------------------------+-----------+------------+
| uc002wwl.3 | chr20 | -      | 30252260 | 30310656 | 30253751 | 30310021 |         3 | 30252260,30309457,30310420, | 30253889,30310151,30310656, | Q07817    | uc002wwl.3 |
| uc002wwn.3 | chr20 | -      | 30252260 | 30310656 | 30253751 | 30310021 |         3 | 30252260,30309457,30310420, | 30253889,30310133,30310656, | Q07817    | uc002wwn.3 |
+------------+-------+--------+----------+----------+----------+----------+-----------+-----------------------------+-----------------------------+-----------+------------+

[donegal77]

Thanks a lot Pierre, I checked the table you sent and I noticed only a difference in the second exon in which the Start is the same while the End is different for 18nt. In your opinion I should work only with the the first sequence uc002wwl.3 - indicated as the main one - or I should take into account also the second one?