We introduce a Stochastic Reaction-Diffusion-Dynamics Model (SRDDM) for simulations of cellular mechanochemical processes with high spatial and temporal resolution. The SRDDM is mapped into the CellDynaMo package, which couples the spatially inhomogeneous reaction-diffusion master equation to account for biochemical reactions and molecular transport within the Langevin Dynamics (LD) framework to describe dynamic mechanical processes. This computational infrastructure allows the simulation of hours of molecular machine dynamics in reasonable wall-clock time. We apply SRDDM to test performance of the Search-and-Capture of mitotic spindle assembly by simulating, in three spatial dimensions, dynamic instability of elastic microtubules anchored in two centrosomes, movement and deformations of geometrically realistic centromeres with flexible kinetochores and chromosome arms. Furthermore, the SRDDM describes the mechanics and kinetics of Ndc80 linkers mediating transient attachments of microtubules to the chromosomal kinetochores. The rates of these attachments and detachments depend upon phosphorylation states of the Ndc80 linkers, which are regulated in the model by explicitly accounting for the reactions of Aurora A and B kinase enzymes undergoing restricted diffusion. We find that there is an optimal rate of microtubule-kinetochore detachments which maximizes the accuracy of the chromosome connections, that adding chromosome arms to kinetochores improve the accuracy by slowing down chromosome movements, that Aurora A and kinetochore deformations have a small positive effect on the attachment accuracy, and that thermal fluctuations of the microtubules increase the rates of kinetochore capture and also improve the accuracy of spindle assembly. Author summary The CellDynaMo package models, in 3D, any cellular subsystem where sufficient detail of the macromolecular players and the kinetics of relevant reactions are available. The package is based on the Stochastic Reaction-Diffusion-Dynamics model that combines the stochastic description of chemical kinetics, Brownian diffusion-based description of molecular transport, and Langevin dynamics-based representation of mechanical processes most pertinent to the system. We apply the model to test the Search-and-Capture mechanism of mitotic spindle assembly. We find that there is an optimal rate of microtubule-kinetochore detachments which maximizes the accuracy of chromosome connections, that chromosome arms improve the attachment accuracy by slowing down chromosome movements, that Aurora A kinase and kinetochore deformations have small positive effects on the accuracy, and that thermal fluctuations of the microtubules increase the rates of kinetochore capture and also improve the accuracy.

Fibrin polymerization involves thrombin-mediated exposure of knobs on one monomer that bind to holes available on another, leading to the formation of fibers. In silico evidence has suggested that the classical A:a knob-hole interaction is enhanced by surrounding residues not directly involved in the binding pocket of hole a, via noncovalent interactions with knob A. We assessed the importance of extended knob-hole interactions by performing biochemical, biophysical, and in silico modeling studies on recombinant human fibrinogen variants with mutations at residues responsible for the extended interactions. Three single fibrinogen variants, yD297N, yE323Q, and yK356Q, and a triple variant yDEK (yD297N/yE323Q/yK356Q) were produced in a CHO (Chinese Hamster Ovary) cell expression system. Longitudinal protofibril growth probed by atomic force microscopy was disrupted for yD297N and enhanced for the yK356Q mutation. Initial polymerization rates were reduced for all variants in turbidimetric studies. Laser scanning confocal microscopy showed that yDEK and yE323Q produced denser clots, whereas yD297N and yK356Q were similar to wild type. Scanning electron microscopy and light scattering studies showed that fiber thickness and protofibril packing of the fibers were reduced for all variants. Clot viscoelastic analysis showed that only yDEK was more readily deformable. In silico modeling suggested that most variants displayed only slip-bond dissociation kinetics compared with biphasic catch-slip kinetics characteristics of wild type. These data provide new evidence for the role of extended interactions in supporting the classical knob-hole bonds involving catch-slip behavior in fibrin formation, clot structure, and clot mechanics.

Microtubules (MTs), a cellular structure element, exhibit dynamic instability and can switch stochastically from growth to shortening; but the factors that trigger these processes at the molecular level are not understood. We developed a 3D Microtubule Assembly and Disassembly DYnamics (MADDY) model, based upon a bead-per-monomer representation of the alpha beta-tubulin dimers forming an MT lattice, stabilized by the lateral and longitudinal interactions between tubulin subunits. The model was parameterized against the experimental rates of MT growth and shortening, and pushing forces on the Dam1 protein complex due to protofilaments splaying out. Using the MADDY model, we carried out GPU-accelerated Langevin simulations to access dynamic instability behavior. By applying Machine Learning techniques, we identified the MT characteristics that distinguish simultaneously all four kinetic states: growth, catastrophe, shortening, and rescue. At the cellular 25 mu M tubulin concentration, the most important quantities are the MT length L, average longitudinal curvature kappa(long), MT tip width w, total energy of longitudinal interactions in MT lattice U-long, and the energies of longitudinal and lateral interactions required to complete MT to full cylinder U-long(add) and U-lat(add) . At high 250 mu M tubulin concentration, the most important characteristics are L, kappa(long), number of hydrolyzed alpha beta-tubulin dimers n(hyd) and number of lateral interactions per helical pitch n(lat) in MT lattice, energy of lateral interactions in MT lattice U-lat, and energy of longitudinal interactions in MT tip u(long). These results allow greater insights into what brings about kinetic state stability and the transitions between states involved in MT dynamic instability behavior.

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.

Fibrin is the major extracellular component of blood clots and a proteinaceous hydrogel used as a versatile biomaterial. Fibrin forms branched networks built of laterally associated double-stranded protofibrils. This multiscale hierarchical structure is crucial for the extraordinary mechanical resilience of blood clots, yet the structural basis of clot mechanical properties remains largely unclear due, in part, to the unresolved molecular packing of fibrin fibers. Here the packing structure of fibrin fibers is quantitatively assessed by combining Small Angle X-ray Scattering (SAXS) measurements of fibrin reconstituted under a wide range of conditions with computational molecular modeling of fibrin protofibrils. The number, positions, and intensities of the Bragg peaks observed in the SAXS experiments were reproduced computationally based on the all-atom molecular structure of reconstructed fibrin protofibrils. Specifically, the model correctly predicts the intensities of the reflections of the 22.5 nm axial repeat, corresponding to the half-staggered longitudinal arrangement of fibrin molecules. In addition, the SAXS measurements showed that protofibrils within fibrin fibers have a partially ordered lateral arrangement with a characteristic transverse repeat distance of 13 nm, irrespective of the fiber thickness. These findings provide fundamental insights into the molecular structure of fibrin clots that underlies their biological and physical properties.

Molecular dynamics has proved itself as a powerful computer simulation method to study dynamics, conformational changes, and interactions of biological macromolecules and their complexes. In order to achieve the best performance and efficiency, it is crucial to benchmark various hardware platforms for the simulations of realistic biomolecular systems with different size and timescale. Here, we compare performance and scalability of a number of commercially available computing architectures using all-atom and coarse-grained molecular dynamics simulations of water and the Ndc80-microtubule protein complex in the GROMACS-2019.4 package. We report typical single-node performance of various combinations of modern CPUs and GPUs, as well as multiple-node performance of the “Lomonosov-2” supercomputer. These data can be used as the practical guidelines for choosing optimal hardware for molecular dynamics simulations. 

Given the need for modern researchers to produce open, reproducible scientific output, the lack of standards and best practices for sharing data and workflows used to produce and analyze molecular dynamics (MD) simulations has become an important issue in the field. There are now multiple well-established packages to perform molecular dynamics simulations, often highly tuned for exploiting specific classes of hardware, each with strong communities surrounding them, but with very limited interoperability/transferability options. Thus, the choice of the software package often dictates the workflow for both simulation production and analysis. The level of detail in documenting the workflows and analysis code varies greatly in published work, hindering reproducibility of the reported results and the ability for other researchers to build on these studies. An increasing number of researchers are motivated to make their data available, but many challenges remain in order to effectively share and reuse simulation data. To discuss these and other issues related to best practices in the field in general, we organized a workshop in November 2018 (https://bioexcel.eu/events/workshop-on-sharing-data-from-molecular-simulations/). Here, we present a brief overview of this workshop and topics discussed. We hope this effort will spark further conversation in the MD community to pave the way toward more open, interoperable, and reproducible outputs coming from research studies using MD simulations.