Computer simulation together with theory and experiments is one of the pillars over which modern science is based. Such simulations shed light on the result of experiments, offer precious insight, replace difficult or expensive experiment, can predict new phenomena.
Our focus is on the study at the atomistic level of condensed systems. In this context simulations can be though of as a form of microscopy that can reveal with the help of computer graphics the behavior of atoms in great microscopic detail.
The advent of ever more powerful computers is of course behind the rapid and pervasive development of computer simulations. However computer power alone cannot meet the need for more complex and accurate simulations. Powerful algorithms are also needed in order to exploit the computer power to the maximum.
Over the years our group has developed a number of widely used methods that have greatly extended the scope of simulation such as the Parrinello-Rahman or the Car-Parrinello method.
More recently we have focused our attention on methods for the study of physical phenomena that occur on very long time scale and that are difficult or impossible to capture in a standard simulations. These methods go under the name metadynamics and variational enhanced sampling. Metadynamics is more than ten years old and it is by now widely used in the community. The Variationally Enhanced Sampling method is much younger but highly promising.
Developing a computational method is however a somewhat dry exercise if performed in abstract. We are driven and inspired by the need to solve real problems in material science, chemistry and biophysics. The list of problems of interest is somewhat long but we can mention two major areas in which we are currently working.
One area is the nucleation and growth of materials. This is done in the contest of a nation-wide coordinated research effort aimed at the discovery of new materials. We also study complex catalytic processes in collaboration with a Swiss chemical industry.
The other major area of application is the study of the proteins and their interaction with ligands where the ultimate goal is to help in the computer design of more effective drugs.