Chapter 4: The Development of Molecular Modelling Programs: The Use and Limitations of Physical Models Check Access

Molecular modelling has become an essential tool for medicinal chemistry, yet even today there are questioning voices as to its value, place and role in drug discovery.¹  This stems in part from the usual cycle of hyperbole, through the well of disillusionment, finally reaching the asymptote of reality; it also comes from a misunderstanding of the strengths and weaknesses of the methodology and hence a lack of ability to correctly interpret the results of modelling experiments and risk-weight them. This essay will look at the evolution of this misconception and will attempt to explain the reason for it. It is true that the fundamental understanding of molecular modelling has not advanced greatly in the last 15 years; the key questions remain the same: molecular representations, thermodynamics, kinetics and the role of solvent. To emphasise this point, several ancient papers will be unashamedly referenced, so that the reader can verify this point for themselves. Proper use of modelling still has much to do with interpretation and knowing the limitations of each technique; there is no magic bullet or workflow of modelling tools that gives the right answer reliably all the time. It follows from this that, unless modellers are deeply embedded in the project teams and the process of design, the impact of modelling will be greatly reduced.