Current Bioinformatics - Volume 5, Issue 4, 2010

Biomacromolecular information is hinged by sequence and structure representations. Because structure is often more conserved than sequence, achieving function inference from structural similarity is easier than from sequence analysis. However, structural information is sparse and only available for a small part of the protein space. Detecting subtle similarities between proteins from sequence depends strongly on the representations used. Continuous-space representations yield promising results in comparative evolution analysis, structural classification and sequence-function/property relationship studies. These simple methods provide a pre-classification and/or feature generation stages to sophisticated classification methods. We review the state-of-the-art in protein sequence graphical representations along with some derived metrics for statistical pattern recognition analysis. In addition, the binding stability pattern of protease-inhibitor complexes is modelled from H-depleted molecular graph representation of protease sequences and ligands using support vector machines with about 80% prediction accuracy.

The post-genomic era, beginning at the end of the Human Genome Project, has led to new needs and challenges in the management of clinical and -omics data. One of the main issues in this new context for biomedical data management is the integration of heterogeneous sources, enabling access to different, remote biological data sources and the interpretation and discovery of new knowledge. Many researchers and practitioners in a wide range of biomedical areas, such as, for instance, all those related to genomic and personalized medicine, have to access these data located at numerous remote sources. Over the last decade, this new scientific context has stimulated research into developing new techniques for seamless web-based data integration and access. Some of the main challenges include the integration of scattered, non-structured public databases, how to deal with sensitive personal information, or how to manage image data. This paper presents a review of methods, techniques and tools for data integration.

Microarray technology has revolutionized biomedical research because it is now possible to concurrently determine the gene expression levels for the whole genome of a target organism. The accuracy of the computed gene expression levels is extremely important for the successful use of this technology. However, microarray gene expression measurements are inherently very ‘noisy’, meaning that appropriate techniques are required to compute accurate gene expression levels. Therefore, the pre-processing of microarray data warrants special consideration. Although there are many candidate techniques for the pre-processing of microarray data, there is no clear-cut best option. In this review, we discuss some of the most important pre-processing techniques applicable to the Affymetrix microarray platform. We also discuss the problems involved in evaluating the different candidate techniques and consider other crucial issues related to the preprocessing of Affymetrix microarray data.

Microarray technology is expanding rapidly providing an extensive as well as promising source of data for better addressing complex questions involving biological processes. The ever increasing number and publicly available gene expression studies of human and other organisms provide strong motivation to carry out cross-study analyses. Besides, microarray technology provides several platforms to investigators that include arrays from commercial vendors like Affymetrix® (Santa Clara, CA, USA), Agilent® (Palo Alto, CA, USA), and other proprietorial arrays of various laboratories. Integration of multiple studies that are based on the same technological platform, or, combining data from different array platforms carries the potential towards higher accuracy, consistency and robust information mining. The integrated result often allows constructing a more complete and broader picture. In this work, we highlight as well as exemplify two frameworks of microarray data integration approaches that are in practice. This follows a discussion on the important issues that may influence any microarray data integration attempt. The review, in general, intends to serve as a starting point for those interested in exploring this area of microarray study, while realizing the pertinent issues underneath.

With the advances in bioinformatics, drug design strategies have been advanced with the focus on target discovery. ‘Proteins and enzymes’ target class represents potential drug target for different diseases. A networking of biochemical pathway of a disease and their drug target needs a thorough understanding. The accessibility of fully sequenced genomes, its total information and their product (proteins and enzymes) have enabled researchers to reconstruct and study the networks based biochemical pathways. Network building of proteins in a biochemical pathway is of utmost importance which helps us to identify the protein (enzyme or receptor) as drug target whose inhibition will achieve to discover a set of compounds (drug like molecules), while incurring minimal side effects. In this paper, we have discussed about the networking of proteins using biochemical pathway which can be used as target for drug development, architecture of biochemical networks, network algorithms, tools and software packages for network building.

Ensemble learning is an intensively studied technique in machine learning and pattern recognition. Recent work in computational biology has seen an increasing use of ensemble learning methods due to their unique advantages in dealing with small sample size, high-dimensionality, and complex data structures. The aim of this article is two-fold. Firstly, it is to provide a review of the most widely used ensemble learning methods and their application in various bioinformatics problems, including the main topics of gene expression, mass spectrometry-based proteomics, gene-gene interaction identification from genome-wide association studies, and prediction of regulatory elements from DNA and protein sequences. Secondly, we try to identify and summarize future trends of ensemble methods in bioinformatics. Promising directions such as ensemble of support vector machines, meta-ensembles, and ensemble based feature selection are discussed.