The gCluster Model: An Approach to Revealing Coregulated Genes Based on General Similarity

Speaker:	Prof. Guoren WANG
		School of Information Science and Engineering
		Northeastern University of China

Title:		"The gCluster Model: An Approach to Revealing Coregulated
		Genes Based on General Similarity"

Date:		Friday, 10 February 2006

Time:		4:40pm - 5:20pm

Venue:		Lecture Theatre H
		(Chen Kuan Cheng Forum, near lift nos. 27/28), HKUST


ABSTRACT:

Currently, pattern-based and tendency-based models are very popular for
clustering co-regulated genes. We propose a new model, g-Cluster, based on
general similarity to mine co-regulated genes. The model is capable of:

(1)	finding co-regulated genes with positive correlation and negative
	correlation,
(2)	mining different categories of micro-array data sets, such as
	Gene-Sample, Gene-Time, and Gene-Sample-Time etc.,
(3)	permitting any magnitude transformation relationship among
	co-regulated genes, including linear and non-linear.

We also design a tree structure and two algorithms with efficient pruning
and optimization strategies to mine all qualified g-Clusters.



**********************
Biography:

Dr. Guoren WANG is a professor of School of Information Science and
Engineering, director of Computer System Institute and head of Computing
Center and Networking Center of Northeastern University, China. His
research interests are bioinformatics, XML data management,
high-dimensional indexing, P2P data management, and parallel data
processing. As visiting scholar, he has visited several universities in
the world, such as Kyushu Univ., Japan, the Chinese University of Hong
Kong, Carleton University, Canada. He published about 60 research papers
in international conferences and journals.