Computational Intelligence for Bioinformatics: The Knowledge Engineering Approach

Abstract

This presentation introduces challenging problems in Bioinformatics (BI) and then applies methods of Computational Intelligence (CI) to offer possible solutions. The main focus of the talk is on how CI can facilitate discoveries from biological data and the extraction of new knowledge.

Methods of evolving knowledge-based neural networks, hybrid neuro-evolutionary systems, kernel methods, local and personalized modeling techniques, characterized by adaptive learning, rule extraction and evolutionary optimization [1], are emphasized among the other traditional CI methods [2].

CI solutions to BI problems such as: DNA sequence analysis, microarray gene expression analysis and profiling, RNAi classification, protein structure prediction, gene regulatory network discovery, medical prognostic systems, modeling gene-neuronal relationship, and others are presented and illustrated.

Fundamental issues in CI such as: dimensionality reduction and feature extraction, model creation and model validation, model adaptation, model optimization, knowledge extraction, inductive versus transductive reasoning, global versus local models, and others are addressed and illustrated on the above BI problems. A comparative analysis of different CI methods applied to the same problems is presented in an attempt to identify generic and specific applicability of the CI methods. A comprehensive environment NeuCom (www.theneucom.com) is used to illustrate the CI methods.

Computational neurogenetic modeling [3],[4] is introduced as a future direction for the creation of new, biologically plausible CI methods for BI and Neuroinformatics applications. These models can help discover patterns of dynamic interaction of genes and neuronal functions and diseases.