Protocol Online logo
Top : Forum Archives: : Bioinformatics and Biostatistics

prediction of putative epitopes - reliable softwares for epitope predictio (Jul/25/2004 )

Hi everyone!
am a new entry in the group.Am doing bioinformatics course in INDIA. I just wanna know about good reliable softwares for epitope prediction.
I have a few here with me but i think it will be a great guidance if anyone of u have already worked on such issue and could give me suitable information as i am not yet sure about how reliable they would be.I tried to relate relaiabity just from the fact that they have been well cited in research ariticles.If anyone has a suggession please let me know .

enjoy life chill size !



Hi there!

There a lots of different methods od in silico prodiction of b-cell and t-cell epitopes, which take into consideration almost every computable information you have. There are methods that compare hydophily and known antigens and know protein structures and so on....

one method is described in

Kolaskar and Tongaonkar (1990); FEBS Letters 276: 172-174 "A semi-empirical method for prediction of antigenic determinants on protein antigens."

it is utilized by a programm call "antigenic". I myself used this method for prediction. I checked for linear and structural epitopes experimentally, too, and what I found was that the computer worls reflected the real world to some degree, but not - not surprisingly - perfectly. E.g. four of five experimentally revealed epitopes were predicted basically correctly, but the fifth real epitope was in a area of the protein where no antigenicty at all was predicted....

So what I wanted to say is, that nothing reflects reality better than reality. Go get yourself a peptide library and check it out! Though maybe the combination of different algorithms may leave you with some useful data.

But that's only my personal oppinion...