Artificial Intelligence Approaches for Knowledge Discovery in Functional Genomics

Project type: Bilateral Collaboration Project
Project duration: 2005 - 2006

Collaborators on project

This is a bilateral project between Faculty of Computer and Information Science and Gad Shaulsky's Lab at Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, Texas.

The goal of this project is to develop and apply new artificial intelligence (AI)-based approaches for knowledge discovery in functional genomics. We will focus on discovery of gene function from experimental genetic data bases, discovery of whole-genome genetic networks from mutant-based observations and discovery of relations between gene’s DNA sequence and its expression. The approaches will stem from abduction, an AI approach that uses background knowledge to provide hypothesis that may explain the experimental data, rule-based machine learning paradigms for discovery of relations between parameters of experimental observations and obtained outcomes, and range of dedicated visualization techniques. Methods developed within the project will be applied to analysis of data on amoeba Dictyostelium discoideum and budding yeast Saccharomyces Cerevisiae. The project will join diverse but complementary research of both partners, each having unrivalled technical competence and track records. The deliverables of the project will include contributions to basic research, components of open-source software for data mining in functional genomics (Orange), and a set of innovative applications.