prediction MHC class I

[odenis]

Hi,

I'm using a site, 

   http://bimas.dcrt.nih.gov/molbio/hla_bind/

in association with papers from Rammensee and from Falk., for the
prediction of CTLs recognized peptides Experimentally I have found 3
peptides on a 35 KD protein. These 3 peptides, and 2 other  were predicted
by this program. Although it seems working, I'm not very glad of the way
results are given (binding score) since some peptides which do not fit the
consensus have a higher score that peptides which have the consensus. Does
anybody else have use this site, what do you think of it, what's the
relevance and do you know other programs?


Thanks

-- 
Olivier DENIS
Laboratory of Mycobacterial Immunology
Pasteur Institute of Brussels, Belgium