Convergent Protocols for Computing Protein–Ligand Interaction Energies Using Fragment-Based Quantum Chemistry

Journal of Chemical Theory and Computation, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 2, 2025

Fragment-based quantum chemistry methods offer a means to sidestep the steep nonlinear scaling of electronic structure calculations so that large molecular systems can be investigated using high-level methods. Here, we use fragmentation compute protein-ligand interaction energies in with several thousand atoms, new software platform for managing fragment-based implements screened many-body expansion. Convergence tests minimal-basis semiempirical method (HF-3c) indicate two-body calculations, single-residue fragments and simple hydrogen caps, are sufficient reproduce obtained conventional supramolecular within 1 kcal/mol at about 1% computational cost. We also demonstrate HF-3c results illustrative trends density functional theory basis sets up augmented quadruple-ζ quality. Strategic deployment facilitates converged biomolecular model alongside high-quality sets, bringing

Language: Английский

---

Quick-and-Easy Validation of Protein–Ligand Binding Models Using Fragment-Based Semiempirical Quantum Chemistry

Journal of Chemical Information and Modeling, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 3, 2025

Electronic structure calculations in enzymes converge very slowly with respect to the size of model region that is described using quantum mechanics (QM), requiring hundreds atoms obtain converged results and exhibiting substantial sensitivity (at least smaller models) which amino acids are included QM region. As such, there considerable interest developing automated procedures construct a based on well-defined criteria. However, testing such burdensome due cost large-scale electronic calculations. Here, we show semiempirical methods can be used as alternatives density functional theory (DFT) assess convergence sequences models generated by various protocols. The these tests reduced even further means many-body expansion. We use this approach examine (with size) protein–ligand binding energies. Fragment-based afford well-converged interaction energies tiny fraction required for DFT Two-body interactions between ligand single-residue acid fragments low-cost way "QM-informed" enzyme size, furnishing an automatable active-site model-building procedure. This provides streamlined, user-friendly constructing binding-site requires neither priori information nor manual adjustments. Extension thermochemical should straightforward.

Language: Английский

---

Prediction of hydration energies of adsorbates at Pt(111) and liquid water interfaces using machine learning

The Journal of Chemical Physics, Journal Year: 2025, Volume and Issue: 162(8)

Published: Feb. 25, 2025

Aqueous phase heterogeneous catalysis is important to various industrial processes, including biomass conversion, Fischer–Tropsch synthesis, and electrocatalysis. Accurate calculation of solvation thermodynamic properties essential for modeling the performance catalysts these processes. Explicit methods employing multiscale modeling, e.g., involving density functional theory molecular dynamics have emerged this purpose. Although accurate, are computationally intensive. This study introduces machine learning (ML) models predict thermodynamics adsorbates on a Pt(111) surface, aiming enhance computational efficiency without compromising accuracy. In particular, ML developed using combination descriptors fingerprints trained previously published water–adsorbate interaction energies, energies solvation, free bound Pt(111). These achieve root mean square error values 0.09 eV 0.04 0.06 demonstrating accuracy within standard modeling. Feature importance analysis reveals that hydrogen bonding, van der Waals interactions, solvent density, together with adsorbate, critical factors influencing thermodynamics. findings suggest can provide rapid reliable predictions properties. approach not only reduces costs but also offers insights into characteristics at Pt(111)–water interfaces.

Language: Английский

---

A focus on delocalization error poisoning the density-functional many-body expansion

Chemical Science, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 1, 2025

Many-body expansions of quantum chemistry data are used for multiscale modeling, but density functional approximations with less than 50% exact Hartree–Fock exchange can bring significant errors (Broderick et al. , Chem. Sci. 2024, https://doi.org/10.1039/D4SC05955G).

Language: Английский

---

Zero-point energies from bond orders and populations relationships

The Journal of Chemical Physics, Journal Year: 2025, Volume and Issue: 162(8)

Published: Feb. 24, 2025

We report two analytical quantum mechanics (QM) models for approximating appropriately scaled harmonic zero-point energies (ZPEs) without Hessian calculations. Following our earlier bond from orders and populations model that takes a similar form as an extended Hückel but uses well-conditioned orbital populations, this work demonstrates proof of concept ZPEs, important component in thermochemistry calculations, while eschewing unfavorably scaling algorithms involving matrices. The ZPE-BOP1 Mulliken hybrid Kohn–Sham density functional theory calculations within Hückel-type defines vibrational energy terms using atom-pairwise parameters are fit to reproduce ZPEs B3LYP more accurate ZPE-BOP2 Hartree–Fock different includes short-range anharmonic term coupled three-body oscillator with seven parameters. Both predict molecules first row elements, outperforms strained long-chain provides competitive those semi-empirical QM methods (e.g., AM1, PM6, PM7, XTB-2) compute This shows progress outlook toward computational use efficiently useful physicochemical properties. It also opportunities approximate would shift traditional bottlenecks away costly such others focus on reliable populations.

Language: Английский

---

Untangling Sources of Error in the Density-Functional Many-Body Expansion

The Journal of Physical Chemistry Letters, Journal Year: 2025, Volume and Issue: unknown, P. 2793 - 2799

Published: March 7, 2025

The many-body expansion provides a framework for data-driven applications of electronic structure theory, including parametrization classical force fields and machine learning. However, we demonstrate that its use significantly amplifies quadrature grid errors when modern density-functional approximations are employed. Standard grids work well in conventional calculations result runaway error accumulation used with the expansion. At same time, delocalization is also exacerbated, leading to exaggerated estimates nonadditive n-body interactions. This illustrated anion–water clusters using SCAN, r2SCAN, ωB97X-V ωB97M-V functionals. By employing dense grids, inherent self-interaction exposed, which can then be mitigated variety other strategies.

Language: Английский

---

Optimal-Reference Excited State Methods: Static Correlation at Polynomial Cost with Single-Reference Coupled-Cluster Approaches

Journal of Chemical Theory and Computation, Journal Year: 2025, Volume and Issue: unknown

Published: April 1, 2025

Accurate yet efficient modeling of chemical systems with pronounced static correlation in their excited states remains a significant challenge quantum chemistry, as most electronic structure methods that can adequately capture scale factorially system size. Researchers are often left no option but to use more affordable may lack the accuracy required model critical processes photochemistry such photolysis, photocatalysis, and nonadiabatic relaxation. A great deal work has been dedicated refining single-reference descriptions ground state via "addition-by-subtraction" coupled cluster pair double substitutions (pCCD), singlet-paired CCD (CCD0), triplet-paired (CCD1), frozen singlet- or amplitudes (CCDf0/CCDf1). By combining wave functions derived from these intermediate representation (ISR), we gain insights into extensibility theory's coverage problem. Our CCDf1-ISR(2) approach is robust face provides enough dynamical accurately predict excitation energies within about 0.2 eV small organic molecules. We also highlight distinct advantages Hermitian ISR construction, avoidance pathological failures equation-of-motion for potential energy surface topology. results prompt us continue exploring optimal theories (excited approaches leverage dependence on initial reference function) potentially economical

Language: Английский

---

Revisiting the Half-and-Half Functional

The Journal of Physical Chemistry A, Journal Year: 2025, Volume and Issue: unknown

Published: April 21, 2025

Hybrid density functionals typically provide significantly better accuracy than semilocal functionals. Conventional wisdom holds that incorporating more 20-25% exact exchange is deleterious to thermochemical properties and should only be used as a last resort, for problems are dominated by self-interaction error. In such cases, the Becke-Lee-Yang-Parr "half-and-half" functional (BH&H-LYP) has emerged go-to choice, especially in time-dependent theory calculations excitation energies. Here, we examine assumption 50% Hartree-Fock sacrifices accuracy. Using sequence of B(α)LYP, with different percentages (0 ≤ α 100), find BH&H-LYP (with = 50) nearly optimal affords similar B3LYP thermochemistry, barrier heights, Although less accurate atomization energies, this emerges sole rationale taboo against values > 25. Overall, reasonable choice error, including charge-transfer complexes core-level While remains valence use appears an acceptable compromise, can without undue concern over its diminished ground-state properties.

Language: Английский

---