An Overview of Biological Data Mining

Introduction

As stated by John Naisbitt, “We are drowning in information but starved for knowledge.” There is a sea change in science due to the tsunami of sequences sweeping over databases. The genome sequencing projects are producing the vast amount of data challenging scientists to potentially reveal the structure and functional relationships of genes and proteins. This phenomenal growth of biological data (biodata) has been witnessed owing to genomic and proteomic technologies such as high-throughput sequencing (HTS) and mass spectrometry, genome-wide two-hybrid screening, DNA microarray, etc. The tremendous amount of sequence information accumulating worldwide have been organized in the form of databanks. However, the data present in the databanks is heterogeneous in nature, such as genomic data, DNA, RNA and protein sequences, protein structures, sequence patterns, sequence annotations, metabolomic data, gene expression data, protein-protein interactions, cross-references, etc. Most of these data are available only on the World Wide Web (Etzioni, 1996). The explosion of these data is impossible to print because it is voluminous and moreover the information in it can only be acquired and assimilated with the aid of data mining tools. With the advent of computers with huge storage capacity and enhanced processing speed, it became easier to use and share information in an electronic format. The digital revolution has made digitized information easy to capture, process, store, distribute and transmit (Fayyad & Uthurusamy, 1996; Inmon, 1996). Besides, the internet allows rapid data transfer and sharing across the globe in an inexpensive manner. The emergence of data mining research such as cluster analysis, outlier analysis, pattern analysis, data visualization and analysis tools contributes to the development of more efficient and scalable methods for knowledge discovery in databanks (Wang et al., 2005). Data mining is a natural evolution of information technology along the path of data collection, database creation, database management, and data analysis and interpretation (Han & Kamber, 2001) in a new field of study known as Bioinformatics. Bioinformatics addresses problems related to the storage, retrieval, and analysis of information about the biological sequence, structure, and function (Altman, 1998).The main idea of biodata mining is to discover knowledge out of sequence, structure and functional information from the web.

Since structural biology has an enormous impact on our understanding of biology and medicine, some of the examples used in this chapter are from EMBL-EBI’s Macromolecular Structure Database (MSD; http://www.ebi.ac.uk/pdbe/), to reflect its close partnership with the wwPDB project. The services and tools name have been changed, but the EBI maintain all existing URLs for external references to the MSD resource. For the 40^thanniversary of PDB (2011), PDBe has turned its attention to a focus on the fundamental problem faced by the structural biology community: “how to make the wealth of structural data available to the larger biomedical community?” (Velankar et al. 2011; Golovin et al. 2004)