| | |
Summary:
A Dynamic Bayesian Approach for
Protein Secondary Structure Prediction
Zafer Aydin1
, Sheila M. Reynolds2
, Jeff A. Bilmes2,3
and William S. Noble1,3
1
Dept of Genome Sciences, 2
Dept of Electrical Engineering, 3
Dept of Computer Science and Engineering,
University of Washington
Abstract
Protein secondary structure prediction is important for accurate prediction of the 3D structure
of a protein. Over the years, many methods have been proposed for predicting secondary
structure. A recently described method that yields stateoftheart performance is the dynamic
Bayesian network (DBN) method of Yao et al. [1], which models the correlation between
neighboring amino acid profiles represented as position specific scoring matrices (PSSMs). In
this work, we replicate the dynamic Bayesian network of Yao et al. [1] using the Graphical
|
