Date |
December 2, 2010 |
Speaker |
Jiuyong Li, Associate professor, School of Computer and Information Science, University of South Australia |
Title |
Using data mining methods to discover regulatory roles of microRNA in Data
|
Abstract |
MicroRNAs (miRNAs) play important roles in many biological processes.
One major role is as a regulator of genes, and dysfunction of miRNA
regulations cause diseases, including cancers. Computational approaches
have been used to determine the functions of miRNAs in data. Previous
work has mainly focused on the identification of miRNAs and their target
genes. However, to understand the regulatory mechanisms of miRNA in
complex cellular systems, it is essential to discover miRNA and mRNA
functional modules involved in complex interactions between miRNAs and
their target genes. In this talk, I will briefly introduce computational
methods for the discovery of miRNA and mRNA regulatory modules, and then
discuss three methods for the discovery developed by us. The first
method discovers miRNA-mRNA regulatory modules by identifying their
associations within specific conditions. The second method uncovers
miRNA-mRNA interactions by Bayesian networks. The third method adapts a
newly developed information retrieval method, Latent Dirichlet
Allocation, to model interactions between miRNAs and mRNAs. All three
methods have been used for real biological data analysis and discovered
some interesting results. Three methods have been published in the
Journal of Biomedical Informatics, BMC Bioinformatics and Bioinformatics.
Bio: Dr Jiuyong Li (John) is an association professor at the School of
Computer and Information Science of University of South Australia. His
research interests are in data mining, biomedical informatics, and data
privacy. He has published a number of papers in leading journals and
conferences in the areas. He has led three projects supported by
Australian Research Council. He has developed a few rule and pattern
discovery methods, one of which has been applied to a number of real
world medical applications. His recent work is mainly on the application
of his rule discovery methods to risk factor discovery and data
inconsistency detection, privacy preserving medical data publishing, and
developing data mining methods for biological applications.
|
|