Provide a computational tool for rapidly estimating ligand binding affinities to biological macromolecules (such as proteins and nucleic acids) using molecular mechanics forcefields and accurate treatments of statistical mechanics
YANK is a code for the rapid computation of small-molecule ligand binding free energies using alchemical free energy methods. YANK uses the OpenMM library that enables rapid molecular mechanics computations on architectures that support OpenCL, which includes GPUs and many multicore CPUs. YANK also intends to simplify these calculations (by encoding current "best practices") so that they might replace other post-docking scoring methods currently in use in the drug design and computational chemistry communities that are less rigorous from a statistical mechanics point of view. YANK also provides a straightforward, extensible platform for exploring new sampling algorithms to further speed these calculations without compromising accuracy.
A C++ version that supports one or more Cuda GPUs through multithreading and implicit solvent calculations with AMBER is currently available.
A Python version that supports OpenCL devices and both explicit and implicit calculations is nearly ready for release. This version is parallelized via MPI.