Goran Krilov

OPLS3: A Force Field Providing Broad Coverage of Drug-like Small Molecules and Proteins

The parametrization and validation of the OPLS3 force field for small molecules and proteins are reported. Enhancements with respect to the previous version (OPLS2.1) include the addition of off-atom charge sites to represent halogen bonding and aryl nitrogen lone pairs as well as a complete refit of peptide dihedral parameters to better model the native structure of proteins. To adequately cover medicinal chemical space, OPLS3 employs over an order of magnitude more reference data and associated parameter types relative to other commonly used small molecule force fields (e.g., MMFF and OPLS_2005). As a consequence, OPLS3 achieves a high level of accuracy across performance benchmarks that assess small molecule conformational propensities and solvation. The newly fitted peptide dihedrals lead to significant improvements in the representation of secondary structure elements in simulated peptides and native structure stability over a number of proteins. Together, the improvements made to both the small molecule and protein force field lead to a high level of accuracy in predicting protein-ligand binding measured over a wide range of targets and ligands (less than 1 kcal/mol RMS error) representing a 30% improvement over earlier variants of the OPLS force field.

Accurate and Reliable Prediction of Relative Ligand Binding Potency in Prospective Drug Discovery by Way of a Modern Free-Energy Calculation Protocol and Force Field

Designing tight-binding ligands is a primary objective of small-molecule drug discovery. Over the past few decades, free-energy calculations have benefited from improved force fields and sampling algorithms, as well as the advent of low-cost parallel computing. However, it has proven to be challenging to reliably achieve the level of accuracy that would be needed to guide lead optimization (∼5× in binding affinity) for a wide range of ligands and protein targets. Not surprisingly, widespread commercial application of free-energy simulations has been limited due to the lack of large-scale validation coupled with the technical challenges traditionally associated with running these types of calculations. Here, we report an approach that achieves an unprecedented level of accuracy across a broad range of target classes and ligands, with retrospective results encompassing 200 ligands and a wide variety of chemical perturbations, many of which involve significant changes in ligand chemical structures. In addition, we have applied the method in prospective drug discovery projects and found a significant improvement in the quality of the compounds synthesized that have been predicted to be potent. Compounds predicted to be potent by this approach have a substantial reduction in false positives relative to compounds synthesized on the basis of other computational or medicinal chemistry approaches. Furthermore, the results are consistent with those obtained from our retrospective studies, demonstrating the robustness and broad range of applicability of this approach, which can be used to drive decisions in lead optimization.

The calculation of transport properties in quantum liquids using the maximum entropy numerical analytic continuation method: Application to liquid para-hydrogen

We present a method based on augmenting an exact relation between a frequency-dependent diffusion constant and the imaginary time velocity autocorrelation function, combined with the maximum entropy numerical analytic continuation approach to study transport properties in quantum liquids. The method is applied to the case of liquid para-hydrogen at two thermodynamic state points: a liquid near the triple point and a high-temperature liquid. Good agreement for the self-diffusion constant and for the real-time velocity autocorrelation function is obtained in comparison to experimental measurements and other theoretical predictions. Improvement of the methodology and future applications are discussed.

Quantum mechanical canonical rate theory: A new approach based on the reactive flux and numerical analytic continuation methods

We present the reactive flux analytic continuation (RFAC) method, based on the quantum reactive flux formalism combined with a numerical analytic continuation approach to calculate quantum canonical rates in condensed phase systems. We express the imaginary time reactive-flux correlation function in terms of a frequency dependent rate constant, and use path integral formalism to derive a working expression suitable for Monte Carlo simulation techniques. The imaginary time data obtained by simulation is analytically continued to the real time using the maximum entropy method to obtain the reaction rate. Motivated by the success of the method to predict the rates for a simple one dimensional parabolic barrier model, we assess its accuracy for a condensed phase reaction modeled by a double-well coupled to a harmonic bath. We note that the method is applicable to a more general Hamiltonian as long as the reaction coordinate can be identified. The reaction rates computed in this fashion are in very good agreem...

Simulation Study of Noncovalent Hybridization of Carbon Nanotubes by Single-Stranded DNA in Water

Recent discovery that single-stranded DNA (ssDNA) binds to carbon nanotubes with high affinity to form soluble hybrids has received great attention as a promising approach to solving the long-standing problem of nanotube solubilization and separation. The mechanism of this process, including the nature of the DNA-nanotube interactions and the molecular structure of the hybrids is still not well understood. Here, we use all-atom replica-exchange molecular dynamics simulations to study the association of several ssDNA decamers with single-walled carbon nanotubes of different chirality in an aqueous environment. The oligonucleotides are found to readily adsorb onto the nanotube surface, after which they undergo a slow structural rearrangement. Cluster analysis of bound DNA conformations as well as population distribution maps computed as a function of several local and global order parameters show that the hybrids exhibit a complex morphology with DNA strands assuming a number of distinct backbone geometries, which depend on both DNA sequence and nanotube diameter. In contrast, the nucleotide bases are found to align parallel to the nanotube surface with a high degree of orientational order. While the binding appears to be primarily driven by energetically favorable pi-stacking of DNA bases onto the nanotube surface, equilibrium distribution of hybrid conformations is modulated by a complex interplay of forces, including the DNA conformational strain and solvent interactions. As a result, the hybrid free-energy landscapes are found to be rugged, with multiple low-lying minima separated by high barriers, several of which are significantly populated at room temperature. Qualitative differences are observed in free energy profiles of purine- and pyrimidine-based oligonucleotide sequences and are attributed to the difference in self-stacking propensity of the bases.

Real time quantum correlation functions. II. Maximum entropy numerical analytic continuation of path integral Monte Carlo and centroid molecular dynamics data

We propose a method which uses centroid molecular dynamics (CMD) [J. Cao and G. A. Voth, J. Chem. Phys. 100, 5106 (1994)] real-time data in conjunction with the imaginary-time data generated using path integral Monte Carlo simulations in a numerical analytic continuation scheme based on the maximum entropy approach. We show that significant improvement is achieved by including short-time CMD data with the imaginary-time data. In particular, for a particle bilinearly coupled to a harmonic bath, these methods lead to significant improvements over previous calculations and even allow accurate determination of transport coefficients such as the diffusion coefficient and mobility for this system. In addition we show how maximum entropy method can be used to extract accurate dynamic information from short-time CMD data, and that this approach is superior to the direct Fourier transform of long-time data for systems characterized by broad, featureless spectral distributions.

Quantum time correlation functions from complex time Monte Carlo simulations: A maximum entropy approach

We present a way of combining real-time path integral Monte Carlo simulations with a maximum entropy numerical analytic continuation scheme in a new approach for calculating time correlation functions for finite temperature many body quantum systems. The real-time dynamics is expressed in the form of the symmetrized time correlation function, which is suitable for Monte Carlo methods, and several simulation techniques are presented for evaluating this function accurately up to moderate values of time. The symmetrized time correlation function is then analytically continued in combination with imaginary time data to obtain the real-time correlation function. We test this approach on several exactly solvable problems, including two one-dimensional systems, as well two cases of vibrational relaxation of a system coupled to a dissipative environment. The computed time correlation functions are in good agreement with exact results over several multiples of the thermal time βℏ, and exhibit a significant improve...

Accurate Binding Free Energy Predictions in Fragment Optimization.

Predicting protein-ligand binding free energies is a central aim of computational structure-based drug design (SBDD)--improved accuracy in binding free energy predictions could significantly reduce costs and accelerate project timelines in lead discovery and optimization. The recent development and validation of advanced free energy calculation methods represents a major step toward this goal. Accurately predicting the relative binding free energy changes of modifications to ligands is especially valuable in the field of fragment-based drug design, since fragment screens tend to deliver initial hits of low binding affinity that require multiple rounds of synthesis to gain the requisite potency for a project. In this study, we show that a free energy perturbation protocol, FEP+, which was previously validated on drug-like lead compounds, is suitable for the calculation of relative binding strengths of fragment-sized compounds as well. We study several pharmaceutically relevant targets with a total of more than 90 fragments and find that the FEP+ methodology, which uses explicit solvent molecular dynamics and physics-based scoring with no parameters adjusted, can accurately predict relative fragment binding affinities. The calculations afford R(2)-values on average greater than 0.5 compared to experimental data and RMS errors of ca. 1.1 kcal/mol overall, demonstrating significant improvements over the docking and MM-GBSA methods tested in this work and indicating that FEP+ has the requisite predictive power to impact fragment-based affinity optimization projects.

On the Bayesian approach to calculating time correlation functions in quantum systems; reaction dynamics and spectroscopy

Abstract We assess the current status, advantages and limitations of the numerical analytic continuation approach to computing time correlation functions in large many-body quantum systems characteristic of condensed phase chemical processes. We determine the quantum correlation function as a function of complex time, and use its analytic properties to select a suitable contour in the complex time plane along which the function can be evaluated efficiently by stochastic simulation methods. The simulation data are then used to obtain the values of the correlation function along the real-time axis through a maximum entropy numerical analytic continuation procedure. This approach is used to compute the dynamical properties of several condensed phase processes including vibrational relaxation lineshapes and canonical reaction rates. We discuss how to improve the accuracy of the numerical analytic continuation methods.

Real time quantum correlation functions. I. Centroid molecular dynamics of anharmonic systems

We investigate the accuracy of the recently proposed centroid molecular dynamics (CMD) method [J. Cao and G. A. Voth, J. Chem. Phys. 100, 5106 (1994)] in the presence of highly anharmonic steep short range repulsive potentials. Such potentials are often present in condensed phases and govern collisions between solvent particles. We compare the results of CMD simulations with exact quantum results for several model one- and two-dimensional nondissipative systems and a one-dimensional system under isobaric conditions. We show that, for nondissipative systems, CMD is accurate only for very short times, and is unable to reproduce the effects of quantum coherences, which play an important role in these few-dimensional systems. CMD gives much better results under isobaric conditions. The correlation functions and the general lineshape of the absorption cross-section in the dipole limit are well reproduced. This is primarily due to dephasing of quantum coherences through inhomogeneous broadening.