Statistical and Mathematical Methods in Bioinformatics- An Overview

Abstract

Bioinformatics is the buzzword these days. Bioinformatics may be considered as the study of information flow within biology and medicine. The first flow is the flow of information from DNA code to biological function. Second flow is the flow of 'in formation in the design and analysis of experiments. Studies in the first flow include methods for sequence alignment, gene finding, RNA expression, protein expression, prediction of protein 3D structure, population genetics, and modelling of genetic network. The second flow begins with a hypothesis (drawn by scanning molecular biology databases), followed by a plan to collect data, execution of an experiment and analysis of the results. Bioinformatics has gained prominence recently because biologists can now collect huge amount of data by using high through put techniques. Large number of relevant databases are available on the internet and a number of software tools and techniques are available for data analysis. Algorithms which have been implemented in these software are based on a number of mathematical and statistical frameworks. Prominent among these are : Dynamic programming, Discriminant analysis, Neural network, Markov model, Multivariate analysis, and it couple of Machine learning techniques.