Kindly note that on-site applications are closed but on-line applications are open until September 1^st at midnight.
 

Molecular dynamics (MD) simulations allow complex phenomena to be studied at a unique spatial and temporal resolution. While conventional simulations provide insights into processes that occur on short time scales, the study of rare events such as protein folding, chemical reactions and phase transitions remains prohibitive in many cases of interest. To extend the time scale that is accessible in MD, several enhanced-sampling methods have been developed over the years. 

Metadynamics was introduced in 2012 as a method to accelerate sampling in MD simulations and at the same time reconstruct free-energy profiles as a function of a small number of descriptors of the system, often known as collective variables (CVs) [1]. Since its introduction, metadynamics has been used to tackle a large variety of problems in computational biology, chemistry, physics, and material science [2,3]. These include problems close not only to basic research in Academia but also to industry, such as applications in the drug discovery field.

Over the past 20 years, there has been active methodological development to extend the capabilities of metadynamics beyond its original formulation [4]. Important milestones in the development of metadynamics include the popular well-tempered variant to improve convergence and tunability of the algorithm, the combination of metadynamics with Replica Exchange Methods to deal with the problem of slow degrees of freedom not described by the CVs used and to allow biasing simultaneously arbitrarily many CVs, and the development of different approaches to recover unbiased thermodynamic and kinetic properties from biased simulations. Strong efforts have also been dedicated to overcoming the challenge of using a small number of CVs to describe complex processes. Notable examples are the so-called Path Collective Variables and the development of machine-learning approaches to automatically detect good CVs that can capture the complexity of biophysical-chemical processes.

Metadynamics and its numerous variants are now implemented in popular software, such as PLUMED [5] and the COLVAR module [6] and they have become indispensable tools in many fields of computational science. Furthermore, over the years a vibrant community has grown around this method. Several events, workshops, and schools have gathered together researchers at every stage of their careers to present recent developments and applications in different fields. Practical hands-on tutorials have been organized, often with the support of CECAM, to train students and postdocs in the use of metadynamics and other enhanced-sampling approaches. This large community has recently put a strong effort in the dissemination of good practices and in the promotion of reproducibility in MD simulations [5].

The objectives of this workshop are to:

• provide an overview of recent methodological advances in the metadynamics field;
• illustrate applications of metadynamics in the fields of computational biology, drug discovery, chemistry, and material science;
• discuss open issues and challenges in the field;
• give the opportunity to students and early-career researchers to discuss their projects in a poster session and contributed talks; 
• promote networking between students, early-career and more experienced researchers, and developers of the metadynamics approach.