Algorithms for mining disease related biological networks

Abstract

A disease can be referred to as the physiological dysfunctionality of an organism. The newlinephysiology of the organism is a collective response of numerous biological processes. newlineThese biological processes are regulated by various bio-molecular interactomes. newlineTypically, diseases are diagnosed and categorized by their clinical manifestation. newlineSometimes, it is challenging to distinguish between the diseases causing similar newlinephenotypic responses or being asymptomatic. Therefore, disease aetiology needs a newlineprecise understanding of the intracellular bio-molecular interaction mechanism that newlinechanges during the disease pathogenesis in an organism. Nowadays, computational newlinenetwork biology has become an indispensable tool for mining multi-scale diseaserelated newlinebiological networks and discovering the connection between genotype and newlinephenotype. In this dissertation, we have developed computational approaches for newlinestudying three types of diseases, viz., genetic disease (Cancer), infectious disease newline(Dengue) and neurodegenerative disorders (Alzheimer s and Parkinson s). Our newlineproposed algorithms utilize several machine learning and graph mining approaches to newlineanalyze the structural and functional properties of the disease-related bio-molecular newlinenetworks responsible for the undesirable physiological states. newline