Kevin Karplus' research aims at predicting protein structure from sequence
information. He uses neural nets for local structure prediction, hidden
Markov models for fold recognition and alignment, and fragment packing
for predicting the structure of new folds. There is a lot to be done at
all levels, but he hopes to be working on new-fold methods in Summer 2003.
Kevin Karplus' home page:
A detailed presentation:
Long list of possible bioinformatics projects (some of which are possible
for undergrads and others) at: