Course overview

Tentitative schedule

May 9, 9:00-11:30: Introduction + integrator exercise

May 9, 13:30-16:00: Force fields

May 10, 9:00-11:30: Tentative: Non-bonded interactions, ensembles

May 10, 13:30-16:00: Tentative: ensembles, water exercise

May 11, 9:00-11:30: Complex models, sampling

May 11, 13:30-16:00: Sampling, free energies

May 12, 9:00-11:30: Hybrid CPU/GPU acceleration & paralllelization (Szilard Pall) + exersize

May 12, 13:30-16:00: Hybrid CPU/GPU acceleration & paralllelization (Szilard Pall) + exersize

May 16, 9:00-11:30: Free-energy exercise

May 16, 13:30-16:00: Systematic coarse graining

Intended learning outcomes

After completion of the course, the student should be able to:

• set up simulations of liquids, polymers and proteins
• analyze structural and dynamic properties
• assess the quality of sampling and be aware of methods to improve sampling
• set up free energy calculations
• apply systematic coarse-graining

Course main content

This course treats advanced molecular dynamics (MD) methodology for classical simulations of liquids, polymers and proteins. MD simulations provide atomistically detailed information on structural and dynamic quantities, but often at a high computational cost. As vibrations of atoms need to be captured, time steps are in the order of femtoseconds, whereas the time scales of interest are often in the microsecond to second range. This has three main implications. Efficient software and hardware is required, using for instance stream computing (on e.g. GPUs). You always need to carefully check the convergence of properties of interest. And if convergence can not be reached, you might need to use methods to improve sampling, such as free energy calculations and coarse-graining. These topic will be covered in lectures, reading of literature and computer labs.

Eligibility

• Basic knowledge of statistical mechanics / thermodynamics

(the course includes a brief introduction to Molecular Dynamics and force fields)

Literature

• D. Frenkel & B. Smit, Understanding Molecular Simulation
• H.J.C. Berendsen, Simulating the Physical World
• Selected articles

Examination

Based on a project using at least one advanced MD technique. You can choose the project, but your plan needs to be approved by the examiner. There is no deadline for handing in the project report, but I would advise not waiting more than a month, otherwise it might never happen.

Requirements for final grade

• Computer Labs using the Gromacs simulation package
• A report on your final project (of sufficient quality)

Offered by

SCI/Applied Physics

Contact

For information and registration, contact Berk Hess <hess@kth.se>

Examiner

Berk Hess <hess@kth.se>