English  |  正體中文  |  简体中文  |  Items with full text/Total items : 90074/105197 (86%)
Visitors : 7167298      Online Users : 128
RC Version 6.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version


    Please use this identifier to cite or link to this item: http://asiair.asia.edu.tw/ir/handle/310904400/3903


    Title: Functional Domain Analysis for Mining Cancer-Related Genes
    Authors: Wu Pei Xuan
    Contributors: Department of Computer Science and Information Engineering
    Keywords: cancer-related gene;functional domain;hierarchical clustering;association rule
    Date: 2003
    Issue Date: 2009-11-18 21:13:05 (UTC+8)
    Publisher: Asia University
    Abstract: Clinically, cancer is a complex family of diseases. From the viewpoint of mo-lecular biology, cancer is a genetic disease resulting from abnormal gene expression due to DNA instability, such as translocation, amplification, deletion or point muta-tions. The purpose of this study is first to analyze the known cancer-related genes, and to discover or predict the rest of unknown cancer-related genes in human genome. To achieve this goal, an approach consisting of three major components is proposed. Firstly, an automatic system has been developed to collect different data sets from the internet, like human genes, proteins, and functional domains. A value-added database is constructed as well. Secondly, the functional domain compositions extracted from proteins are adopted for grouping the associated cancer-related genes into a number of clusters by hierarchical clustering and association rule methods. For each cluster of genes, there exist some functional domains in common. That is to say, the common functional domains represent the characteristics or rules of a cluster of genes. Finally, these rules are compared with the entire human genes for cancer gene prediction. PubMed literature search is conducted to verify roughly the correctness of the predic-tive result. The grouping result shows that the 167 known cancer-related genes can be categorized into four classes. The first and second classes contain 55 genes, and each gene possesses unique functional domain composition in this population. The third class comprises 24 genes grouped into 11 small clusters, and the genes in each cluster have one unique functional domain. The fourth class includes 88 genes grouped into 26 clusters, and the genes in each cluster have some functional domains in common. The approached is based on functional domain analysis of the cancer-related genes to generate a set of rules for prediction of another set of genes that are also cancer-related. An assessment indicated that a total accuracy of 81.45% was reached. In comparison with all human genes, totally 2578 genes satisfy the rules from the func-tional analysis of the known cancer-related genes. The result shows that the approach is effective yet efficient in cancer-related gene prediction. Future work will focus on the influence of weighted functional domain composition in cancer gene prediction.
    Appears in Collections:[資訊工程學系] 博碩士論文

    Files in This Item:

    File SizeFormat
    0KbUnknown236View/Open


    All items in ASIAIR are protected by copyright, with all rights reserved.


    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback