基于人工神经网络和小波分析技术的基因表达谱数据分析-电路与系统专业论文.docxVIP

AbstractDNA Abstract DNA microarrays can be used to measure the expression Ievels of thousands of genes simultaneously and so it is usually an effectire tooI of cancer research Gene expression data offer potentiaI insight into gene function and regulatory mechanisms and aid in better understandina of carcinogenesis．DNA microarrays have been used in many research fields，such as cancer classification and sub．classification。gene function and co—regulation research and about how to produce more effective drugs，etc． Firstly，we discussed wavelet analysis based denoise method．It reaches a good denoising effective．And then we propose the ioint use of discrete wavelet transform and correlation coefficients ranking method to build a feature extractor．Some test and comparison experiments have been made to evaluate the performance of the proposed feature extraction scheme．and jt has been proved that：high classification rate，robustness，not sensitive to the number of features．and can be used in most classification system． After feature extraction， we research two application fields： cancer classification and gene clustering．In cancer classification。we studied probabilistic neural network(PNNl mainly．We evaluate the performance using PNN on three public data sets：Golub’s Acute Leukemia data set． BhatIachariee’s adenocarcinoma data set and AIon。s colon cancer data set． Correct rate higher than most of traditionaI methods has been obtained． The task of discovering every gene’s functionality may not be as daunting as it sounds due to the surprising fact that there is a high level of similarity between the genes and chromosomes of what were previously thought to be very different organisms．We make use of clustering method to assist the research about gene function and co—regulated genes． Keywords Feature Extraction；Classification；Clustering；Wavelet Analysis；Artificial Neural Network；Gene Expression 独创性声明本人声明所呈交的论文是我个人在导师指导下进行的研究工作及取得的研 独创性声明 本人声明所呈交的论文是我个人在导师指导下进行的研究工作及取