Development of algorithm for gene expression analysis through MaxEnt based multivariate information theory: MaxEnt based algorithm for analyzing of gene expression

Abstract

Human Genome Project generates a large amount of data. Simultaneously, it opens up Functional Genomics that ultimately results towards the development of Systems Biology. This subject area mostly relies on automated high-throughput technologies for data capture from biological systems. High-throughput technology generates a huge amount of data of multiple variables simultaneously. So handling and analysis of big data in this area becomes a challenge. For systems level functional annotations with designing of circuitry of the biological system, analysis of gene expression data holds the central position of Systems Biology research. To handle and analysis of these vast amount of data, generally different clustering and soft computing methodologies are already employed. Due to empirical nature of biological data Information Theoretic analysis has not been employed for many years in biological/medical problems, though it is well employed for physical systems. However, availability and application of automated techniques for biomedical data capture, recent time Information Theoretic approach is employed. This area of research can be accelerated further if some automated computational algorithm is available. To address this we have developed an algorithm for Multivariate Information Theoretic analysis. Output from our developed algorithm successfully tallies with the results available in literature.