Efficient functional bioinformatics tools: towards understanding biological processes

Experimental high-throughput techniques in biology are generating large amounts of data related to genes and proteins at different levels. The functional analysis of such datasets is a necessary key step in their interpretation. In this contribution we review different methodologies and applications for functional bioinformatics in the context of automated data and text analysis in biology. Introduction The post-genomics era has been characterized by the emergence of several high-throughput techniques such as DNA microarrays, protein chips, chip-on-chip, and more recently, the new generation sequencers [1-4], which are allowing the scientific community to study biological processes from a global perspective to understand the basis of development and diseases. Nevertheless, the great potential of these technologies is resulting in a problem when researchers have to deal with the vast amounts of data that are generated by these technologies. Life sciences in general are generating huge amounts of data with the accelerated development of high-throughput technologies. Therefore, in parallel to the accumulation of the experimental information, there is an obvious need to analyse this huge amount of data. The challenge lies not only in the data-processing area, in which the bioinformatics community has made significant advances, but also in the interpretation of the experimental data. An essential task in this analysis is to translate gene signatures into information that can assist in understanding the underlying biological mechanisms. During the last few years, the Functional Bioinformatics Group from the National Center for Biotechnology in Madrid (http://bioinfo.cnb. csic.es) has focused on the development of several methods and tools to assist researchers in the analysis and interpretation of genomics and proteomics data. One of our main goals has been to provide the research community with easyto-access and easy-to-use tools, implementing complex data-analysis methodologies that allow a more complete understanding of cellular processes and diseases. In this way, we have generated a set of web-based applications that addresses three main fields: gene-expression data analysis, functional interpretation of large lists of genes or proteins and automatic knowledge extraction from the biomedical literature. In our effort to develop applications for a broad number of users able to deal with computing-intensive problems, we have designed these tools incorporating efficient implementations of the algorithms and their variants, most of them using parallel and grid computing. In addition, most of these tools can be accessed via web-services, which allow users to integrate them in their analysis workflows. Applications The mission of our research group is the development and implementation of algorithms that facilitate the understanding of biological processes through the application of statistical and data-mining techniques. Figure 1 shows an overview of the applications and databases that we have developed, can be divided into three different groups: