Novel Hybrid Neural Networks for 'In Silico' Metabolic Modeling

H. Richardson, L. Macklem, and M. Zohdy (USA)

Keywords

Neural Networks, Metabolic Pathways

Abstract

Neural networks are effective intelligent computational tools; simultaneously there is a recent need for better metabolic simulation programs which handle recent gene expression information from genomes of sequenced mi croorganisms. This paper describes a novel suite of hybrid Neural Network programs, which are designed to model metabolic pathways due to initial concentrations and envi ronmental inputs. Five features were added in order to opti mize these neural programs. They include: polymorphism, inheritance, encapsulation, context, and hints of available metabolic biological knowledge. With these key features accuracy in the program predicted outputs is increased to +/0.02. Therefore, the goal of optimizing hybrid neural programs was highly successful.

Important Links:



Go Back