Correlating Human and Mouse Unigene clusters

Robert Xuequn Xu xux at its.caltech.edu
Thu Jul 30 12:44:39 EST 1998


Human Unigene info was from ftp://ftp.ncbi.nlm.nih.gov/, The version
was "UniGene Build #42"; the mouse version was "UniGene Build #21".
In NCBI's repository/unigene directory, the files are called
"Hs.data" and "Mm.data". The file format is in kind of
"object-oriented" style, and I wrote a Perl program to convert each
of them into two relational tables. The matching was done by joining
mouse and human tables on the "GENE" field - mouse and human
unigenes with the same gene names were retrieved.

So now you see the result is a limited set of homologous genes
between them (those that have the same GENE symble). Other possible
homologous genes were ignored (those that do not yet have gene
names, or have different gene names).

Xuequn Xu


On Wed, 29 Jul 1998, David Kristofferson wrote:

> Xuequn,
> 
> Yes, this is just what I was looking for.  Thank you!  What version of
> Unigene did you use and how did you do the matching?
> 
> Dave Kristofferson
> 





More information about the Comp-bio mailing list