29 August – 2 September 2016, Lorentz Centre, Leiden, NL
Short description: The 5 day workshop in the E-CAM classical molecular simulation work-package brought together 31 participants, and focused on the identification of reaction coordinates and the construction of reliable models from atomistic simulation data, such as nucleation at phase transitions, drug binding, protein-protein interactions, protein folding, association, and self-assembly. The computational methodologies discussed are relevant to fields ranging from physics and chemistry, to materials science and molecular biology. Several issues were discussed and partially resolved, including for instance, how to use machine learning to find collective variables and reaction coordinates. A series of community needs for methods and software tools were identified.
12 – 14 September 2016, Cranage Hall, UK
Short description: The 3 day workshop in the E-CAM electronic structure work-package brought together 23 participants. Density functional theory is now a mature technology that is increasingly finding application industrially. Industrial use demands accuracy, which is now being addressed by validation and verification exercises such as the D Codes Project. A recurring theme of the meeting was the importance of having well specified error bars, and clear definitions of applicability of the different methods. It also was clear that dynamics, multi-scale modelling and finite temperature effects remain challenging topics. To expand the applicability of DFT, and post DFT theories, advanced techniques from computer science are already being applied and can be expected to play an increasing role, these include not only the use of massively parallel hybrid CPU/GPU hardware, but also efficient strategies based on machine learning.
6 – 10 June 2016, EPF Lausanne, CH
Short description: The 5 day workshop in the E-CAM quantum dynamics work-package brought together 40 participants. New methodologies or major enhancement of existing methods of quantum dynamics were identified including: ring polymer molecular dynamics at the cost of classical molecular dynamics, which paves the way for the inclusion of quantum effects in condensed phase reactions, and in biological processes. Quantum dynamical simulations are increasingly relevant to industry, including hardware design (e.g. coherence and interference effects for quantum control or design of q-buts), pharmaceuticals (e.g. tunneling in enzymatic reactions), and energy production or storage. In the context of E-CAM, new simulation methods and algorithms for quantum computing are being developed with IBM, with similar collaborations involving other participants. Surface hopping and multiple spawning are methods of choice to simulate excited-state dynamics of molecular systems. Applications of these techniques to dyes and emitters have been reported, and brought insight to the design of molecules for various applications such as dye solar cells (e.g. collaborative projects with Dyesol, BASF) or organic light emitting diodes (with BASF, Novaled). Excited state, and in particular non-adiabatic dynamics is potentially interesting also for pharmaceutical companies, for example, to prevent photo damage leading to skin cancer.
16 – 20 May 2016, UCD, Dublin
Short description: This workshop, which attracted 32 participants from across Europe, focused on parallel programming models, emerging HPC platforms and a range of scalable molecular dynamics and electronic structure codes. Among the lecturers there were representatives from the DL_POLY, CP2K and GROMACS, Quantum Espresso, SIESTA and ASE developer communities as well as PRACE application performance experts.