Prediction of DNA Binding Transcription Factor segments under Specified structure

• Main Authors: Jen-Jay Hsu, 徐振傑
• Other Authors: Yen-Jen Oyang
• Format: Others
• Language: zh-TW
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/46288973583523923659

碩士 === 國立臺灣大學 === 資訊工程學研究所 === 96 === This thesis discusses the design of a predictor aimed at identifying the secondary structures in a transcription factor that are involved in interaction with the DNA. In particular, the design of the predictor has been optimized for identifying the alpha-helix structures involved in interaction with the DNA due to their prevalence. In the design of the predictor, the support vector machine (SVM) was employed and the study reported in this thesis focused on the features exploited for making prediction. In the experiments conducted in this study, two datasets have been used. The first dataset was derived from the TF-DNA complexes deposited in the Protein Data Bank (PDB) and the second dataset was derived from the TF sequences deposited in SWISS-PROT. With respect to identifying the alpha-helix structures involved in interaction with the DNA, the predictor proposed in this thesis delivered sensitivity of 75%, precision of 80%, and specificity of 92% with the first dataset and sensitivity 65%, precision 85%, and specificity 98% with the second dataset.