Henrik Fredholm

A novel approach to prediction of the 3-dimensional structures of protein backbones by neural networks

Three‐dimensional structures of protein backbones have been predicted using neural networks. A feed forward neural network was trained on a class of functionally, but not structurally, homologous proteins, using backpropagation learning. The network generated tertiary structure information in the form of binary distance constraints for the Cα atoms in the protein backbone. The binary distance between two Cα atoms was 0 if the distance between them was less than a certain threshold distance, and 1 otherwise. The distance constraints predicted by the trained neural network were utilized to generate a folded conformation of the protein backbone, using a steepest descent minimization approach.

Neural Networks Applied to Protein Structure

Artificial neural networks are cognitive computational strategies which involve training on well‐established examples rather than programming of the traditional sequential type. They are already enjoying considerable success with the prediction of protein secondary structures, and preliminary results with tertiary structures, though less impressive, nevertheless provide grounds for optimism.