Feature Selection and Classification of Microarray Data with Integrated Approaches

Abstract

Biological data are data or measurements collected from biological sources, which are often stored or exchanged in a digital form. Examples of biological data are DNA base-pair sequences, and population data used in ecology. These data consists of information from various researches concerned with the field of genomics, proteomics, metabolomics, microarray gene expression and phylogenetics (Ahmad et al., 2011). These databases store data related to gene function, structure, localization (both cellular and chromosomal), clinical effects of mutations as well as similarities of biological sequences and structures. newlineThe number of these types of datasets being generated is increasing at a phenomenal rate worldwide (Reichhardt, 1999). For example, as of April 2014, the GenBank repository of nucleic acid sequences contained 17,17,44,486 entries and is estimated to double every 18 months (http://www.ncbi.nlm.nih.gov/genbank/statistics) and the Swiss-Prot database of protein sequences containing 5,45,388 sequence entries, comprising 19,39,48,795 amino acids abstracted from 2,28,536 references (http://web.expasy.org/docs/relnotes/relstat.html). In addition, the H. influenza genome (Fleischmann et al., 1995) released complete sequences for over 40 organisms, ranging from 450 genes to over 1,00,000. The number of biological datasets is still increasing owing to the increasing number of studies on gene expression, protein structure and their interactions.