"A KNOWLEDGE-BASED EXPERT SYSTEM IN BIOINFORMATICS: AN APPLICATION TO REVERSE ENGINEERING GENE REGULATORY NETWORK"

The huge amount of biological data has spread the development of plenty of bionformatics tools, databases and web services. In order to face a computational biology problem, there not exist only a way, but different methodologies and strategies, with their own pros and cons, can be applied. In this PhD thesis I present a knowledge-based expert system that aims at helping a bionformatics researcher in the choice of the proper strategy and heuristic in order to resolve a bioinformatics issue. The Knowledge Base of the system is structured by means of an ontology and codes the expertise about the application domain. KB is organized into decision-making modules that introduce a set of metareasoning levels. The proposed expert system is the core reasoning component of BORIS (Bionformatics Organized Resources - an Intelligent System) framework, a research project High Performance Computing and Networking Institute of National Research Council (ICAR-CNR). BORIS, based on a hybrid architecture, can be seen as a crossover between Decision Support System and Workflow Management System because it not only provides decision support, but it help the User in the proper configuration and running of algorithms, tools and services implementing the suggested strategies and, at the same time, builds a workflow that traces both the decision-making activity and the execution of tasks and tools. The whole system will be applied to an actual case study: the reverse engineering of Gene Regulatory Network.