High-throughput electronic biology: mining information for drug discovery

In the past 10 years, the life sciences have seen a proliferation of electronic data, emerging from systems such as databases, text-mining technologies, high-throughput techniques and

'omics' platforms (for example, DNA microarray).

In this article, we review some of the recent developments in the field of electronic biology (eBiology), which uses these resources as a substrate for new drug discovery. As extensive

reviews on data sets and tools already exist, we highlight how these resources can be applied directly to solve bottlenecks in the industry.

A number of approaches to the application of eBiology in drug discovery are identified, ranging from deep 'systems biology' to 'project-specific' analyses and focus on high-throughput

techniques.

A set of examples are given, which look at the power of applying multiple resources simultaneously to build layers of evidence and end with the identification of novel drug targets. In these

scenarios, the expert in that disease area is a key partner, without which the exercise is unlikely to succeed.

Although there are an increasing number of examples of target mining in the literature, there is also a need to consider how one then translates these biological hypotheses into drug

discovery programmes. To this end, we consider the data sources and techniques that can be used to apply a business focus to the results of this mining.

As hypotheses turn into real programmes, we consider further workflows to support these later stages of discovery, looking at issues such as druggability, selectivity and understanding the

action of a compound both in vitro and in vivo. Although these areas are traditionally addressed by computational chemists, there is much to be gained by the use of techniques and resources

familiar to eBiologists.

Last, we discuss the future needs in eBiology and how the area must be primarily led through scientific creativity, rather than technical considerations.

The vast range of in silico resources that are available in life sciences research hold much promise towards aiding the drug discovery process. To fully realize this opportunity,

computational scientists must consider the practical issues of data integration and identify how best to apply these resources scientifically. In this article we describe in silico

approaches that are driven towards the identification of testable laboratory hypotheses; we also address common challenges in the field. We focus on flexible, high-throughput techniques,

which may be initiated independently of 'wet-lab' experimentation, and which may be applied to multiple disease areas. The utility of these approaches in drug discovery highlights the

contribution that in silico techniques can make and emphasizes the need for collaboration between the areas of disease research and computational science.

The authors would like to thank T. Turi, J. Lanfear, S. Campbell, I. Harrow, J. Keeling and the Pfizer PharmaMatrix team for their valuable insight and support while preparing this article.

We also wish to gratefully acknowledge the input and suggestions from the reviewers. Finally, we wish to acknowledge the contribution of those who develop and maintain a vast landscape of in

silico resources and apologize to those who we have been unable to cite in this article.

Computational Biology Group, Pfizer, Groton, Connecticut, USA

A term used to describe experiments or experimental results that are held electronically.

Controlled compendium of vocabularies across the spectrum of life sciences.

The US National Library of Medicine's controlled vocabulary used for indexing articles for Medline.

A specific location in a DNA sequence at which different people can have a different DNA base. Differences in a single base could change the protein sequence, leading to disease (for

example, sickle-cell disease), or have no known consequences.

The testing of novel therapeutic strategies (in humans) that were developed through basic laboratory experimentation. Observations taken 'from the bedside to the bench' also constitute

translational medicine.

RSS is a family of data formats used to publish frequently updated digital content, such as news feeds and alerts from many web sites on the internet.

(QSAR). Mathematical relationships linking chemical structure and pharmacological activity in a quantitative manner for a series of compounds. Methods that can be used in QSAR include

various regression and pattern-recognition techniques.

A biological assay that assesses DNA damage caused by small molecule drugs in bacterial cells.

Anyone you share the following link with will be able to read this content: