1. Could you explain the significance of your article to the non-specialist?

In the 21st century, airplanes and other complex machines are designed almost entirely using computers.   By contrast, drug design is still largely a process of trial and error.   This is partly because the physics underlying the action of drugs is very complicated.   Advances in computer power and simulation algorithms are enabling the development of new methods for computer-aided drug discovery.   The methods we describe provide a more physically realistic treatment of the binding of small molecules to macromolecular drug targets and address key challenges that have limited progress in this field, especially the treatment of water.  

2 What has motivated you to conduct this work?

I felt that the field of high-throughput docking had largely stagnated over the past 5-10 years, in part because of an over-reliance on empiricism, as opposed to exploiting our understanding of the physics of protein-ligand binding.   A physics-based approach makes it possible to systematically advance the field and take advantage of further increases in computing power.  

3. Where do you see this work developing in the future?

Computer power will continue to increase exponentially for the near future, which will enable even more physically realistic models of protein-ligand binding.   We are particularly enthusiastic about the potential to use methods of quantum mechanics to provide dramatically more accurate descriptions of key forces involved in ligand binding.  

4. Are there any particular challenges facing future research in this area?

There are many.   The most important is that there are many aspects to the physics, and improving the treatment of one aspect (e.g., the effects of water) does not necessarily lead to a dramatic improvement in accuracy if other aspects are neglected (e.g., entropic losses).   That is, there is no "magic bullet" that will suddenly propel the field forward.   It will take work on many fronts simultaneously before inhibitor discovery can become an engineering discipline.