Support Vector Machine-based Prediction and Classification of
Coiled-Coil Structure

This thesis extends work undertaken by Jamie Feinberg '03 and
Prof. Mona Singh. Presently, Support Vector Machines have extremely
high success rates in separating two-stranded coiled-coil proteins
from proteins without coiled-coil structures. The aim of the thesis is
to extend these techniques to 3- and 4-stranded coiled-coils, and to
produce a classifier that identifies and groups 2-, 3-, and 4-
stranded coils.

Joseph Barillari