Issue No. 02 - March/April (2002 vol. 17)
<p>The authors provide a brief overview of the application of machine-learning methods to proteomic problems. They outline a novel strategy for the complete prediction of protein 3D coordinates. The strategy relies on three main successive stages: prediction of structural features, topology, and actual coordinates.</p>
contact map, evolutionary information, protein structure prediction, protein contacts, recurrent neural networks, secondary structure, solvent accessibility
P. Baldi and G. Pollastri, "A Machine-Learning Strategy for Protein Analysis," in IEEE Intelligent Systems, vol. 17, no. , pp. 28-35, 2002.