DESCRIPTION OF COURSES
BI 524 TOOLS AND TECHNIQUES FOR BIOLOGICAL DATA MINING (2L+1P) I
To understand various algorithms of machine learning approaches.
Quality of Biological Data & Data Accuracy; General issues regarding Biological Databases: Representation of errors due to (machines, 3D structural and sequence data of proteins and nucleic acid, Proteomics and Micro array data).
Optimization Techniques: Steepest Descent, Conjugate Gradient, Newton-Raphson, Simulated annealing in Biomolecular Structure Optimization; Genetic Algorithms: Ab initio methods for structure prediction; Lattice, SOM, etc., Information theory, entropy and relative entropy, Stochastic Grammars & natural languages processing techniques.
Clustering and Classification Algorithms: Hierarchical and non-hierarchical Clustering, K-Means clustering, Grid based clustering, Analysis of MD trajectories, Protein Array data Analysis.
Dynamic Programming and application in bioinformatics: Sequence Alignments, Structure Alignments; Foundations for Machine learning Techniques: Hidden Markov Model, Neural Network, Bayesian modeling, The Cox-Jaynes Axiomes; Support Vector machine & Ant colony optimization: Multiple Sequence Alignments, Biomolecular Structure Prediction; Fuzzy logic system & application in bioinformatics; Introduction to WEKA package; Clustering and classifications, Protein Array data Analysis.