More about HKUST
Machine Learning Approaches Towards System Level Understanding of Gene Regulation and Gene Function
Speaker: Dr. Renqiang Min Yale University Title: "Machine Learning Approaches Towards System Level Understanding of Gene Regulation and Gene Function" Date: Monday, 4 April 2011 Time: 4:00pm - 5:00pm Venue: Lecture Theatre F (near lifts 25/26), HKUST Abstract: With the availability of high-throughput genomics, functional genomics, and proteomics data during the last decade, elucidating the underlying mechanisms of gene regulation and gene function at a system level becomes possible, which helps to understand the molecular mechanisms of human diseases. In this talk, I will present machine learning approaches to predicting gene regulation and gene function at different biological levels. First, at gene expression and protein product level, I will show how hierarchical Bayesian graphical models were used to infer the post-transcriptional regulatory mechanisms of small non-coding RNAs called microRNAs, by integrating sequence data, expression data, and proteomics data. Second, at protein sequence level, I will show how the functions of genes can be predicted directly from protein sequences in SCOP, using learned random-walk kernels and learned empirical-map kernels. Last, at cellular morphology level, I will present how kernel methods and deep embedding methods based on undirected graphical models were constructed to characterize novel functions of yeast essential genes, by analyzing the microscopy image data from genome-wide high-content screening of yeast temperature-sensitive mutants. ******************* Biography: Renqiang (Martin) Min is a postdoctoral associate working with Prof. Mark Gerstein in Program of Computational Biology and Bioinformatics, Yale University. He received his Master and PhD degrees in Computer Science from Machine Learning Group, Department of Computer Science, University of Toronto, and obtained his Bachelor degree in Computer Science from Nankai University. His research focuses on bioinformatics and machine learning, in particular, personal genomics, sequence analysis, gene regulation inference, gene function prediction, kernel methods, graphical models, and deep learning.