Computer-aided molecular modelling: research study or research tool?

Developments in computational methods and equipment have produced a new type of research chemist, who prefers to calculate properties as well as measure them, either to gain a better understanding of microscopic molecular behaviour per se, or to guide a broader scientific study using a so-called 'rational' approach. While there is good reason to believe that significant results can be obtained this way, it is clear that only some of the 'tools of the trade' are sufficiently robust to present to those who are not experts in the field. This paper discusses the underlying basis for molecular modelling techniques, describing their history, state of maturity and prospects for future development. The intention is to outline the scope that these play in an industrial research environment, and to examine how they can successfully be incorporated as routine research tools. Computer-based molecular modelling is not a unique discipline. Some comparisons are drawn between it and other computer-based techniques that have reached a greater degree of maturity, in order to highlight the points made.