Disulfide connectivity prediction using secondary structure information and diresidue frequencies
Speaker: Fabrizio Ferre , Boston College
Date: October 18 2004
Time: 11:30AM to 1:00PM
Location: TOC Lab at MIT, Stata Center, 32-G575
Host: P Clote/ BC & Bonnie Berger/ MIT- Math & CSAIL
Contact: Kathleen Dickey, 617 253-3037, kvdickey@mit.edu
Relevant URL: http://www-math.mit.edu/compbiosem/Cysteine sulfhydryl groups can form strong covalent bonds (the disulfide bonds) that stabilize the tertiary and/or quaternary structures of proteins. Knowing the disulfide connectivity of a protein chain can be useful for the understanding of the protein functions and may help protein folding prediction algorithms. We describe a stand-alone software to predict disulfide bond partners in a protein
given only the amino acid sequence, using a novel neural network architecture (the
diresidue neural network), given input of symmetric flanking regions of N- and Cterminus
half-cystines augmented with residue secondary structure (helix, coil, sheet)
as well as evolutionary information. The approach is motivated by the observation of
a bias in the secondary structure preferences of free cysteines and half-cystines, and
by the good performance obtained using diresidue position specific scoring matrices.
As calibrated by ROC curves from 4-fold cross-validation, our conditioning on
secondary structure leads to a performance comparable or improved over the current
state-of-the-art method. A slight drop in performance is seen when secondary
structure is predicted rather than derived from three dimensional protein structures. A
web engine (http://clavius.bc.edu/~clotelab/DiANNAforDBC/)
allows the on-line use of this tool.
See other events that are part of Bioinformatics Seminar Series 2004/2005
See other events happening in October 2004
