This course focuses on data representation methods and comparative and knowledge extraction algorithms for massive biological data. Topics include pairwise sequence alignment, dynamic programming, multiple sequence alignment, phylogenetic tree estimation, approximated methods, sequential motif representation, rapid homology search techniques versus large-scale database, protein structure modeling and structure prediction, and so on. No biological nor biocheical knowledge are prerequisite. The basic biology notions are introduced within the course, and students are required to consider the topics in the view of computational algorithms and its complexity.
Biological information analysis is significantly important for our society in the 21st century in order to improve our quality of life, environment, and safety. Thus this course is aiming at providing the fundamental understanding to the nature of biological data and typical algorithms, like as dynamic programming, repeatedly used in this area. On the other hand, most of the methods explained in the course are also applicable to wide range of engineering subjects. We aim to provide this course to students as an illustrative example how computer science techniques are applied in a specific real-world problem.
By the end of this course, students will be able to:
1) Explain several data representation for sequence analysis (e.g. regular expression, profile matrix, HMM).
2) Explain the notion and implementation of dynamic programing, as well as its several applications in bioinfomatics.
3) Explain the important role of approximated methods in multiple sequence alignment and phylogenetic tree estimation, in terms of computational complexity.
4) Explain the notion of e-value and p-value in homology search against a large database, and compute the values.
5) Explain several algorithmic techniques to make faster homology search against a large database, and
6) Explain protein tertiary structure model and structure prediction methods.
biological information, algorithm, complexity, dynamic programming, hidden Markov model, gene, protein
|Intercultural skills||Communication skills||Specialist skills||Critical thinking skills||Practical and/or problem-solving skills|
Each class starts from explanation of new topic (through notion, example, systems, applicational importance, etc.). At the end of class, students are given exercise problems related to the lecture given that day to solve.
|Course schedule||Required learning|
|Class 1||Biological information and its importance - Computational view of, gene, genome, protein, cell, and body||Understand hierarchical system, scale comparison, and information flow|
|Class 2||Global sequence alignment - Optimal path search, dynamic programmig, global sequence alignment||Calculate global sequence alignment based on dynamic programming|
|Class 3||Local sequence alignment - Protein, amino acids, local sequence alignment||Calculate local sequence alignment based on dynamic programming|
|Class 4||Multiple sequence alignment - Complexity of multiple alignment, heuristic methods||Calculate multiple sequence alignment based on star method or tree-based method|
|Class 5||Phylogenetic tree estimation －Distance matrix method，character state method，bootstrap evaluation||Calculate phylogenetic tree based on UPGMA method or NJ method|
|Class 6||Homology search against database －Amino acid mutation matrix，hit significance，e-value，bit score, p-value||Calculate e-value and p-value for a hit in homology search|
|Class 7||Faster methods for homology search －FASTA，BLAST，PSI-BLAST||Build a k-mer index table for faster similarity search|
|Class 8||Motif representation and extraction －Regular expression, profile matrix, hidden Markov model||Understand several representation methods for sequence motif|
|Class 9||Probabilistic modeling for sequence analysis －Coding region prediction, Markov model||Understand roles of probabilistic modeling for sequence analysis|
|Class 10||RNA secondary structure prediction －RNA secondary structure，Nussinov algorithm，Zuker method||Calculate RNA secondary structure prediction based on dynamic programing|
|Class 11||Genome-wide sequence analysis －Requirement for further speed-up, BLAT，RMAP||Understand approximate methods for rapid sequence analysis|
|Class 12||Protein secondary structure prediction －Protein secondary structure，DSSP code，neural net，PSI-PRED||Understand protein secondary structure prediction methods|
|Class 13||Protein tertiary structure prediction －Homology modeling, fold recognition, fragment assembly method||Understand protein tertiary structure prediction methods|
|Class 14||Gene expression and gene regulartory network estimation －Gene expression，DNA microarray，gene regulatory network||Understand gene regulatory network estimation methods|
|Class 15||Chemical compound metabolism and metabolic network estimation －Compound metabolism，metabolic network||Understand metabolic pathway network estimation methods|
Original slides by Akiyama and Yamamura are provided.
(Ed. Japanese Society of Bioinformatics). Introduction to Bioinformatics. Tokyo: Keio University Press; ISBN：978-4-7664-2251-1. (Japanese)
Students' knowledge of data representations, algorithms, and applications in biological information analysis, and their ability to apply them to problems will be assessed.
Midterm and final exams 80%, exercise problems 20%.