臺灣博碩士論文加值系統

生物資訊的資料，隨著「人類基因體計劃」的啟動日漸增加，然而有不少基因體與蛋白質體已經定序，卻未知其功能，因此預測功能的方法有其重要性。
蛋白質的功能與其結構有著密切的關係。本篇論文中，將蛋白質結構考慮至其中，發展出一個序列結構排序法來預測蛋白質功能。這個方法是利用蛋白質一級結構序列及其二級結構序列來建造出一個HMM數學統計模型，接著利用已知的蛋白質資料訓練模型，再將所得的機率值拿來做為預測時的依據。此篇論文是依著名的Gene Ontology分類來做為實驗中分類的依據。蛋白質資料則是從Protein Data Bank取得。
依據序列結構排序法的原則，提出兩個預測的系統。第一個系統是假設某未知蛋白質用不同類別的模型來計算整體的期望值，並以此值來判斷該蛋白質屬於那個類別，其精確率為63%。然而蛋白質可能有一種或是多種的功能，而第二個系統則再建立不是此類別的數學模型，不但可以測試出某未知蛋白質所可能存在的一種或多種功能，而且其精確率提昇為81%。

As the Human Genome Project (HGP) progresses, there are more biological data available. Many proteins and genes have been sequenced, but their functions remain unknown. Therefore, function predicting methods become important.
It is recognized that molecular structure of a protein is closely related to its function. In this thesis, we develop sequence-structure alignment method to predict protein function. This method uses protein sequence and secondary structure to build a statistical HMM model which is trained according to pre-existing databases. In this paper, the function is classified by Gene Ontology (GO), and the protein data files are collected from Protein Data Bank (PDB).
Two predicted algorithms are adopted. The first one builds different HMM models based on function classification of GO, and then determines overall probability of protein belonging to this function. The accuracy of this algorithm reaches to 63%. However, the protein may have one or more functions. The second algorithm builds additional HMM model using the data not belong to this category. This algorithm can predict multiple functions of an unknown protein. Its accuracy is raised to 81%.

Table of Contents
Abstract in Chinese i
Abstract ii
Acknowledgment iii
Table of Contents iv
List of Figures vii
List of Tables ix
Chapter 1 Introduction 1
1-1 Significance of protein 1
1-2 Motivation and purpose 2
1-3 Dissertation organization 3
Chapter 2 Research Background 4
2-1 The overview of protein structure 4
2-1-1 Primary structure 4
2-1-2 Secondary structure 6
2-1-3 Tertiary structure 8
2-1-4 Quaternary structure 8
2-2 Methods to predict protein function 9
2-2-1 Sequence alignment 9
2-2-2 Phylogenetic profiles 13
2-2-3 Artificial neural networks 14
2-3 Gene ontology 15
2-4 Hidden markov model 17
Chapter 3 Sequence-Structure Alignment Method 20
3-1 Data set 20
3-2 Training and predicting 21
3-2-1 Using hidden markov model 22
3-2-2 Calculate every variable's frequency of passing through 25
3-2-3 Calculate variable’s probability 27
3-2-4 Probability translates into log value 28
3-2-5 Predicting method 30
3-2-6 An example for predicting 31
3-3 Test 32
Chapter4 Performance of Our Approach 36
4-1 Data source and our database 36
4-2 Training 37
4-3 Performance 40
Chapter5 Conclusion and Future Work 44
Reference 45
Appendix A 48

[1] Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) “Basic local alignment search tool”, J. Mol. Biol., 215(3):403-410.
[2] Pearson, William R (2000) “Flexible sequence similarity searching with the FASTA3 program package”, Methods Mol Biol 132, 185–219.
[3] M. Pellegrini, E.M. Marcotte, M.J. Thompson, D. Eisenberg, and T.O. Yeates (1999) “Assigning protein functions by comparative genome analysis: Protein phylogenetic profiles”, Proc. Natl. Acad. Sci. USA, vol. 96, pp. 4285-4288.
[4] T.F. Smith, M.S. Waterman (1981) “Identification of common molecular subsequences”, J. Mol. Biol. 147, 1, pp. 195-197.
[5] S. B. Needleman and C. D. Wunsch (1970) “Ageneral method applicable to the search for similarities in the amino acid sequence of two proteins”, J. Mol. Biol. 147: 195-197.
[6] T. F. Simth and M. S. Waterman (1981) “Comparison of biosequences”, Adv. Appl. Math. 2: 482-489.
[7] The website has described LCS
http://en.wikipedia.org/wiki/Longest-common_subsequence_problem
[8] Barton, G. J. (1993c) “An efficient algorithm to locate all locally optimal alignments between two sequences allowing for gaps”, Comput. Appl. Biosci. 9, 729-734.
[9] Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ. (1997)“Gapped BLAST and PSI-BLAST: a new generation of protein database search programs”, Nucleic Acids Res ,25(17):3389-3402.
[10] J. Zhang and T. L. Madden (1997) “PowerBLAST: a new network BLAST application for interactive or automated sequence analysis and annotation”, Genome Res, Vol. 7, No. 6, pp. 649-656.
[11] NCBI BLAST
http://www.ncbi.nlm.nih.gov/BLAST/
[12] Lars Juhl Jensen (2002) “Prediction of Protein Function from Sequence Derived Protein Features”, Center for Biological Sequence Analysis, BioCentrum-DTU，Technical University of Denmark, Lyngby.
[13] The Gene Ontology
http://www.geneontology.org/index.shtml
[14] Ashburner, M., C. A. Ball, et al. (2000) “Gene ontology: tool for the unification of biology. The Gene Ontology Consortium”, Nature Genetics 25(1): 25-29.
[15] Lawrence R. Rabiner (1989) “A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition. Proceedings of the IEEE”, 77 (2), p. 257–286.
[16] C. Yan, D. Dobbs, V. Honavar and D. Dobbs. (2004) “A two-stage classifier for identification of protein-protein interface residues”, Bioinformatics, 20(S1):i371--i378.