John Overington & Coauthors on Effective Drug Targets
Emerging Research Fronts Commentary, August 2011
|
Article: Opinion - How many drug targets are there?
Authors: Overington, JP;Al-Lazikani, B;Hopkins,
AL |
John P. Overington, Bissan Al-Lazikani, & Andrew Hopkins talk with ScienceWatch.com and answer a few questions about this month's Emerging Research Front paper in the field of Pharmacology & Toxicology.
Why do you think your paper is highly
cited?
The sequence of the human genome was complete, and there had been irrational exuberance over the applications of the genome sequence, and other large-scale technologies, such as high-throughput screening (HTS), structural genomics, and combinatorial chemistry. However, signposts were there that pointed to falling industry productivity, as opposed to an explosion of new and novel therapies. Surprisingly, to us, there was little available data, either on the molecular structures of drugs, or their mechanisms of action and molecular targets, that could be analyzed to make predictions of the optimal strategies for target discovery, to aid drug rescue and repositioning, and so forth.
In this work, we identified the efficacy targets of the drugs (the target believed to be responsible for the therapeutic effect) as opposed to proteins that the drug just happened to bind to, and classified them in terms of function and subcellular location. It is clear that drugs interact with many proteins in the body, sometimes with positive functional effects (polypharmacology), sometimes causing adverse events, and more often with no significant effect. Separating out key therapeutically significant interactions was key.
Does it describe a new discovery, methodology, or
synthesis of knowledge?
Coauthor Bissan Al-Lazikani
Coauthor Andrew Hopkins
The work not only brought together, in a structured and consistent way, knowledge about the previous success of the pharmaceutical industry, but also mapped out the molecular pharmacology of historical drug targets and identified common features among historically successful targets. The key step was in assembling the data, which was broadly scattered over both the primary literature and data sources such as Prescribing Information documents. In hindsight, it would have been good to have developed a stronger taxonomy for drug types and their mechanisms (which we have now done).
Would you summarize the significance of your paper
in layman's terms?
The paper highlights the fact that current drugs, of which there are around 1500, work therapeutically through about 250-300 distinct molecular targets. The number of targets is significant in that it represents just 1-2% of the genes in the human genome. The pharmaceutical industry is currently in a productivity crisis, with far fewer drugs being launched than needed to support the companies at their current size. Understanding the properties of historical "good" drug targets will be essential in future development of novel ones.
How did you become involved in this research, and
how would you describe the particular challenges, setbacks, and
successes that you've encountered along the way?
We had all been long interested in the molecular aspects of drug discovery, and had the good fortune to have worked across multiple disciplines—including structural biology, bio- and chemo-informatics, and computational data-mining. We had also worked together for a long time, and knew the difficulty in assembling the data needed. The paper was part of a broad theme of developing new informatics methods to make drug discovery more efficient and effective.
We now work at three different institutes, the European Bioinformatics Institute, the University of Dundee, and the Institute of Cancer Research. It is interesting, to us at least, that we have all left industry positions and feel we can make more of a difference in the academic sector, where there is a lot of interest in building chemical biology and drug discovery capabilities.
Where do you see your research leading in the
future?
We have a broad range of interests, but all focused on improving drug discovery through computational approaches. We see that Open Data—freely accessible, downloadable, and usable without restriction as being key; and also that these data provide a solid platform for innovation and the development of IP—getting the informatics infrastructure right is crucial. Other areas of our current research include exploration of drug safety issues, network pharmacology, data integration, and biological drugs.
Do you foresee any social or political
implications for your research?
Hopefully, as a community, we will use the data generated in this work,
together with similar analytical and integrative approaches, to learn how
to translate the huge volumes of (-omics) data we have invested in and
collected, into tangible healthcare products. This is particularly
important given the downsizing and changes in pharmaceutical company
strategies and investments.
John P. Overington, C.Chem.
Computational Chemical Biology
European Bioinformatics Institute (EMBL-EBI)
Wellcome Trust Genome Campus
Hinxton, Cambridge, United Kingdom
Bissan Al-Lazikani
Institute of Cancer Research
Sutton, Surrey, United Kingdom
Andrew Hopkins
Biological Chemistry and Drug Discovery
School of Life Sciences
University of Dundee
Dundee, United Kingdom
ADDITIONAL INFORMATION:
- View a poster from (C) Macmillan Publishers Ltd (2006) describing this paper.
KEYWORDS: CURRENT DRUGS, MOLECULAR TARGETS, CONSENSUS NUMBER, TYROSINE KINASE, THERAPEUTIC TARGETS, GENOME, EXPLORATION, INHIBITION, DISCOVERY, DATABASE, PROGRESS, BINDING.