交通大学、清华大学 统计学研究所 专题演讲 题　目：A Pattern Recognition Approach to Infer Genetic Interactions 主讲人：谢淑蓉博士(中央研究院 统计科学研究所) 时　间：97年10月31日(星期五)上午10:40-11:30 (上午10:20-10:40茶会於交大统计所428室举行) 地　点：交大综合一馆427室 Abstract For any time course microarray data in which the gene interactions and the associated paired patterns are dependent, the proposed pattern recognition approach (PARE) can infer time-lagged genetic interactions, a challenging task due to the small number of time points and large number of genes. PARE utilizes a nonlinear score to identify subclasses of gene pairs with different time lags. In each subclass, PARE extracts nonlinear characteristics of paired gene expression curves and learns weights of the decision score applying an optimization algorithm to microarray gene expression data (MGED) of some known interactions, from biological experiments or published literature. For both transcriptional compensation (TC) and transcriptional interactions (TIs) in yeast, PARE outperforms a time-lagged correlation approach and the latest advance in graphical Gaussian models. Further, several predicted TC/TD interactions coincide with existing pathways involving Sgs1, Srs2 and Mus81. This reinforces the possibility of applying genetic interactions to predict pathways of protein complexes. We further apply PARE to a human obese project. Using known 70 TIs and MGED of human adipocytes-derived cell lines as training, we predict several genetic interactions of interest; all the predicted TIs are consistent with existing results from other studies, while some predicted interactions have biological bearings. This is a joint work (published in Bioinformatics, 2008) with Cheng-Long Chuang, Chih-Hung Jen and Chung-Ming Chen. 敬 请 公 布 欢 迎 参 加 --

※ 发信站: 批踢踢实业坊(ptt.cc) ◆ From: 140.113.114.158