Hao Jiang

Work place: Advanced Modeling and Applied Computing Laboratory Department of Mathematics The University of Hong Kong, Hong Kong, China

E-mail: haohao@hkusuc.hku.hk


Research Interests: Computer systems and computational processes, Solid Modeling, Data Structures and Algorithms, Mathematics of Computing


Hao Jiang got her B. Sc. in computational mathematics from Harbin Institute of Technology (2009). Currently she is a Ph.D. student in the Department of Mathematics, the University of Hong Kong. Her research interest is mathematical modeling and scientific computing.

Author Articles
Kernel Techniques in Support Vector Machines for Classification of Biological Data

By Hao Jiang Wai-Ki Ching Zeyu Zheng

DOI: https://doi.org/10.5815/ijitcs.2011.02.01, Pub. Date: 8 Mar. 2011

In this paper, we consider the problem of protein classification, which is a important and hot topic in bioinformatics. We propose a novel kernel based on the KSpectrum Kernel by incorporating physico-chemical and biological properties of amino acids as well as the motif information for the captured protein classification problem. Similarity matrix is constructed based on an AAindex2 substitution matrix which measures the amino acid pair distance. Together with the motif content posing importance on the protein sequences, a new kernel is then constructed. We adopt the Eigen-matrix translation techniques for improving the classification accuracy. Experimental results indicate that the string-based kernel in conjunction with SVM classifier performs significantly better than the traditional spectrum kernel method. Furthermore, numerical examples also confirm the use of the Eigenmatrix translation techniques as general strategy.

[...] Read more.
On Construction of Gene-PDB Structure Mapping with Applications in Functional Annotation of Human Genes

By Xi Chen Hao Jiang Wai-Ki Ching Limin Li

DOI: https://doi.org/10.5815/ijitcs.2011.02.08, Pub. Date: 8 Mar. 2011

Protein 3D structure is one of the key factors in recognizing gene functions. The availability of protein structure data in Protein Data Bank (PDB) enables us to conduct gene function analysis. However, the molecules in the PDB, whose structures have been determined, are always not corresponding to a unique gene. That is to say, the mapping from gene to PDB is not one-to-one. Thus this uncertain property complicates the analysis and increases the difficulty of gene function analysis. In this paper, we attempt to tackle this challenging issue and we study the problem of predicting gene function from protein structures based on the gene-PDB mapping. We first obtain the gene-PDB mapping, which is important in representing a gene by the structure set of all its corresponding PDB molecules. We then define a new gene-gene similarity measurement based on the structure similarity between PDB molecules. We further show that this new measurement matches with gene functional similarity nicely. This means that the measurement we introduced here can be useful for gene function prediction. Numerical examples are given to demonstrate our claim.

[...] Read more.
Other Articles