Michael S. Sellers

A coarse-grain force field for RDX: Density dependent and energy conserving

We describe the development of a density-dependent transferable coarse-grain model of crystalline hexahydro-1,3,5-trinitro-s-triazine (RDX) that can be used with the energy conserving dissipative particle dynamics method. The model is an extension of a recently reported one-site model of RDX that was developed by using a force-matching method. The density-dependent forces in that original model are provided through an interpolation scheme that poorly conserves energy. The development of the new model presented in this work first involved a multi-objective procedure to improve the structural and thermodynamic properties of the previous model, followed by the inclusion of the density dependency via a conservative form of the force field that conserves energy. The new model accurately predicts the density, structure, pressure-volume isotherm, bulk modulus, and elastic constants of the RDX crystal at ambient pressure and exhibits transferability to a liquid phase at melt conditions.

Development of Bacterial Display Peptides for use in Biosensing Applications

Recent advances in synthetic library engineering continue to show promise for the rapid production of reagent technology in response to biological threats. A synthetic library of peptide mutants built off a bacterial host offers a convenient means to link the peptide sequence, (i.e., identity of individual library members) with the desired molecular recognition traits, but also allows for a relatively simple protocol, amenable to automation. An improved understanding of the mechanisms of recognition and control of synthetic reagent isolation and evolution remain critical to success. In this paper, we describe our approach to development of peptide affinity reagents based on peptide bacterial display technology with improved control of binding interactions for stringent evolution of reagent candidates, and tailored performance capabilities. There are four key elements to the peptide affinity reagent program including: (1) the diverse bacterial library technology, (2) advanced reagent screening amenable to laboratory automation and control, (3) iterative characterization and feedback on both affinity and specificity of the molecular interactions, and (3) integrated multiscale computational prescreening of candidate peptide ligands including in silico prediction of improved binding performance. Specific results on peptides binders to Protective Antigen (PA) protein of Bacillus anthracis and Staphylococcal Enterotoxin B (SEB) will be presented. Recent highlights of on cell vs. off-cell affinity behavior and correlation of the results with advanced docking simulations on the protein-peptide system(s) are included. The potential of this technology and approach to enable rapid development of a new affinity reagent with unprecedented speed (less than one week) would allow for rapid response to new and constantly emerging threats.

XPairIt Docking Protocol for peptide docking and analysis

The mechanics of peptide–protein docking has long been an area of intense interest to the computational community. Here we discuss an improved docking protocol named XPairIt which uses a multitier approach, combining the PyRosetta docking software with the NAMD molecular dynamics package through a biomolecular simulation programming interface written in Python. This protocol is designed for systems where no a priori information of ligand structure (beyond sequence) or binding location is known. It provides for efficient incorporation of both ligand and target flexibility, is HPC-ready and is easily extensible for use of custom code. We apply this protocol to a set of 11 test cases drawn from benchmarking databases and from previously published studies for direct comparison with existing protocols. Strengths, weaknesses and areas of improvement are discussed.

In silico design of smart binders to anthrax PA

The development of smart peptide binders requires an understanding of the fundamental mechanisms of recognition which has remained an elusive grail of the research community for decades. Recent advances in automated discovery and synthetic library science provide a wealth of information to probe fundamental details of binding and facilitate the development of improved models for a priori prediction of affinity and specificity. Here we present the modeling portion of an iterative experimental/computational study to produce high affinity peptide binders to the Protective Antigen (PA) of Bacillus anthracis. The result is a general usage, HPC-oriented, python-based toolkit based upon powerful third-party freeware, which is designed to provide a better understanding of peptide-protein interactions and ultimately predict and measure new smart peptide binder candidates. We present an improved simulation protocol with flexible peptide docking to the Anthrax Protective Antigen, reported within the context of experimental data presented in a companion work.

Shock simulations of a single-site coarse-grain RDX model using the dissipative particle dynamics method with reactivity

In discrete particle simulations, when an atomistic model is coarse-grained, a tradeoff is made: a boost in computational speed for a reduction in accuracy. The Dissipative Particle Dynamics (DPD) methods help to recover lost accuracy of the viscous and thermal properties, while giving back a relatively small amount of computational speed. Since its initial development for polymers, one of the most notable extensions of DPD has been the introduction of chemical reactivity, called DPD-RX. In 2007, Maillet, Soulard, and Stoltz introduced implicit chemical reactivity in DPD through the concept of particle reactors and simulated the decomposition of liquid nitromethane. We present an extended and generalized version of the DPD-RX method, and have applied it to solid hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX). Demonstration simulations of reacting RDX are performed under shock conditions using a recently developed single-site coarse-grain model and a reduced RDX decomposition mechanism. A description of the...

Exponential-six potential scaling for the calculation of free energies in molecular simulations

An adjustable, scaled form of the exponential-six (exp-6) potential is presented. The potential form allows stable scaling from a fully interacting exp-6 system to a non-interacting reference system for the direct computation of free energy differences or efficient particle growth simulations, particularly for high-density systems. Additional scaling parameters were introduced to overcome known endpoint effects, whereby reducing the potential to an ideal gas state can produce singularities in simulation averages or prohibit the sampling of close particle distances. The scaled potential is validated in several ways, using Hamiltonian thermodynamic integration, by comparison to vapour–liquid and solid–liquid coexistence free energies reported in the literature, and by the application of the Gibbs–Helmholtz equation. Forms of the scaled exp-6 potential for its implementation into molecular simulations and the thermodynamic integration methods are also developed.

Prediction of protein-peptide interactions: application of the XPairIt API to anthrax lethal factor and substrates

As software and methodology develop, key aspects of molecular interactions such as detailed energetics and flexibility are continuously better represented in docking simulations. In the latest iteration of the XPairIt API and Docking Protocol, we perform a blind dock of a peptide into the cleavage site of the Anthrax lethal factor (LF) metalloprotein. Molecular structures are prepared from RCSB:1JKY and we demonstrate a reasonably accurate docked peptide through analysis of protein motion and, using NCI Plot, visualize and characterize the forces leading to binding. We compare our docked structure to the 1JKY crystal structure and the more recent 1PWV structure, and discuss both captured and overlooked interactions. Our results offer a more detailed look at secondary contact and show that both van der Waals and electrostatic interactions from peptide residues further from the enzymes catalytic site are significant.