FEATURE SELECTION IN HEXAMPLOT TO ASSESS DRUG EFFECT IN CDNA MICROARRAY EXPERIMENTS

摘要：

Three-color cDNA microarray experiments are designed to assess drug effects on a genomic scale in an original way.With this kind of expression data, we propose an effective algorithm, named HoughFeature, to extract the significant features of polymorphic gene expressions to quantify drug effects in hexaMplots.The Hough technique is used in our algorithm to detect the featured lines in hexaMplots corresponding to the diverse levels of drug effects on differentially expressed genes.Thus, based on hexaMplots, the side and therapeutic effects of drugs can be quantified with our methodology.We apply the framework to the experimental microarray data to assess the complex effect of PW-1 (an extract of Chinese medicine) on TCDD toxified HepG2 cells in detail.Such a methodology may be useful in forefront gene therapy to predict disease susceptibility, implement drug therapy, and assess their effects.

关键词： HoughFeature Three-color cDNA microarray Gene expression HexaM plot Hough transform

作者: HONG-YA ZHAO HONG YAN

作者单位: Department of Electronic Engineering, City University of Hong Kong, Kowloon, Hong Kong Department of Electronic Engineering, City University of Hong Kong, Kowloon, Hong Kong;School of Ele

会议类型: 国际会议

会议名称: 2007 International Conference on Machine Learning and Cybernetics(IEEE第六届机器学习与控制论国际会议)

会议地点: 香港

会议语种:英文

页码: 2202-2207

在线出版日期: 2007-08-19（万方平台首次上网日期，不代表论文的发表时间）