The introduction of accelerator devices such as graphics processing units (GPUs) has had profound impact on molecular dynamics simulations and has enabled order-of-magnitude performance advances using commodity hardware. To fully reap these benefits, it has been necessary to reformulate some of the most fundamental algorithms, including the Verlet list, pair searching, and cutoffs. Here, we present the heterogeneous parallelization and acceleration design of molecular dynamics implemented in the GROMACS codebase over the last decade. The setup involves a general cluster-based approach to pair lists and non-bonded pair interactions that utilizes both GPU and central processing unit (CPU) single instruction, multiple data acceleration efficiently, including the ability to load-balance tasks between CPUs and GPUs. The algorithm work efficiency is tuned for each type of hardware, and to use accelerators more efficiently, we introduce dual pair lists with rolling pruning updates. Combined with new direct GPU–GPU communication and GPU integration, this enables excellent performance from single GPU simulations through strong scaling across multiple GPUs and efficient multi-node parallelization.

Structural determination of N- and O-linked glycans as well as polysaccharides is hampered by the limited spectral dispersion. The computerized approach CASPER, an acronym for computer assisted spectrum evaluation of regular polysaccharides, uses liquid state NMR data to elucidate carbohydrate structure based on agreement with predicted (1)H and (13)C chemical shifts. We here demonstrate developments based on multiple through-bond J-based correlations that significantly enhance the credence to the sequence connectivities proposed in the analysis exemplified by an oligosaccharide and a bacterial polysaccharide. The approach is also suitable for predicting (1)H and (13)C NMR chemical shifts of synthesized oligosaccharides and glycoconjugates, thereby corroborating a proposed structure.

Understanding and predicting permeability of compounds through skin is of interest for transdermal delivery of drugs and for toxicity predictions of chemicals. We show, using a new atomistic molecular dynamics model of the skin's barrier structure, itself validated against near-native cryo-electron microscopy data from human skin, that skin permeability to the reference compounds benzene, DMSO (dimethyl sulfoxide), ethanol, codeine, naproxen, nicotine, testosterone and water can be predicted. The permeability results were validated against skin permeability data in the literature. We have investigated the relation between skin barrier molecular organization and permeability using atomistic molecular dynamics simulation. Furthermore, it is shown that the calculated mechanism of action differs between the five skin penetration enhancers Azone, DMSO, oleic acid, stearic acid and water. The permeability enhancing effect of a given penetration enhancer depends on the permeating compound and on the concentration of penetration enhancer inside the skin's barrier structure. The presented method may open the door for computer based screening of the permeation of drugs and toxic compounds through skin.

In the present study we have analyzed the molecular structure and function of the human skin's permeability barrier using molecular dynamics simulation validated against cryo-electron microscopy data from near native skin. The skin's barrier capacity is located to an intercellular lipid structure embedding the cells of the superficial most layer of skin - the stratum corneum. According to the splayed bilayer model (Iwai et al., 2012) the lipid structure is organized as stacked bilayers of ceramides in a splayed chain conformation with cholesterol associated with the ceramide sphingoid moiety and free fatty acids associated with the ceramide fatty acid moiety. However, knowledge about the lipid structure's detailed molecular organization, and the roles of its different lipid constituents, remains circumstantial. Starting from a molecular dynamics model based on the splayed bilayer model, we have, by stepwise structural and compositional modifications, arrived at a thermodynamically stable molecular dynamics model expressing simulated electron microscopy patterns matching original cryo-electron microscopy patterns from skin extremely closely. Strikingly, the closer the individual molecular dynamics models' lipid composition was to that reported in human stratum corneum, the better was the match between the models' simulated electron microscopy patterns and the original cryo-electron microscopy patterns. Moreover, the closest-matching model's calculated water permeability and thermotropic behaviour were found compatible with that of human skin. The new model may facilitate more advanced physics-based skin permeability predictions of drugs and toxicants. The proposed procedure for molecular dynamics based analysis of cellular cryo-electron microscopy data might be applied to other biomolecular systems.

Free energy calculation has long been an important goal for molecular dynamics simulation and force field development, but historically it has been challenged both by limited performance, accuracy, and creation of topologies for arbitrary small molecules.This has made it difficult to systematically compare different sets of parameters to improve existing force fields, but in the last few years several authors have developed increasingly automated procedures to generate parameters for force fields such as Amber, CHARMM, and OPLS. Here, we present a new framework that enables fully automated generation of GROMACS topologies for any of these force fields and an automated setup for parallel adaptive optimization of high-throughput free energy calculation by adjusting lambda point placement on the fly. We have calculated solvation free energies of * To whom correspondence should be addressed † Royal Institute of Technology ‡ Stockholm University 1 50 different small molecules using the GAFF, OPLS-AA and CGenFF force fields and four different water models, and by including the often neglected polarization costs we show that the common charge models are somewhat underpolarized.