TY - JOUR
T1 - Introducing KICK-MEP
T2 - exploring potential energy surfaces in systems with significant non-covalent interactions
AU - García-Argote, Williams
AU - Ruiz, Lina
AU - Inostroza, Diego
AU - Cardenas, Carlos
AU - Yañez, Osvaldo
AU - Tiznado, William
N1 - Publisher Copyright:
© The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2024.
PY - 2024/11
Y1 - 2024/11
N2 - Context: Exploring potential energy surfaces (PES) is fundamental in computational chemistry, as it provides insights into the relationship between molecular energy, geometry, and chemical reactivity. We introduce Kick-MEP, a hybrid method for exploring the PES of atomic and molecular clusters, particularly those dominated by non-covalent interactions. Kick-MEP computes the Coulomb integral between the maximum and minimum electrostatic potential values on a 0.001 a.u. electron density isosurface for two interacting fragments. This approach efficiently estimates interaction energies and selects low-energy configurations at reduced computational cost. Kick-MEP was evaluated on silicon-lithium clusters, water clusters, and thymol encapsulated within Cucurbit[7]uril, consistently identifying the lowest energy structures, including global minima and relevant local minima. Methods: Kick-MEP generates an initial population of molecular structures using the stochastic Kick algorithm, which combines two molecular fragments (A and B). The molecular electrostatic potential (MEP) values on a 0.001 a.u. electron density isosurface for each fragment are used to compute the Coulomb integral between them. Structures with the lowest Coulomb integral are selected and refined through gradient-based optimization and DFT calculations at the PBE0-D3/Def2-TZVP level. Molecular docking simulations for the thymol-Cucurbit[7]uril complex using AutoDock Vina were performed for benchmarking. Kick-MEP was validated across different molecular systems, demonstrating its effectiveness in identifying the lowest energy structures, including global minima and relevant local minima, while maintaining a low computational cost.
AB - Context: Exploring potential energy surfaces (PES) is fundamental in computational chemistry, as it provides insights into the relationship between molecular energy, geometry, and chemical reactivity. We introduce Kick-MEP, a hybrid method for exploring the PES of atomic and molecular clusters, particularly those dominated by non-covalent interactions. Kick-MEP computes the Coulomb integral between the maximum and minimum electrostatic potential values on a 0.001 a.u. electron density isosurface for two interacting fragments. This approach efficiently estimates interaction energies and selects low-energy configurations at reduced computational cost. Kick-MEP was evaluated on silicon-lithium clusters, water clusters, and thymol encapsulated within Cucurbit[7]uril, consistently identifying the lowest energy structures, including global minima and relevant local minima. Methods: Kick-MEP generates an initial population of molecular structures using the stochastic Kick algorithm, which combines two molecular fragments (A and B). The molecular electrostatic potential (MEP) values on a 0.001 a.u. electron density isosurface for each fragment are used to compute the Coulomb integral between them. Structures with the lowest Coulomb integral are selected and refined through gradient-based optimization and DFT calculations at the PBE0-D3/Def2-TZVP level. Molecular docking simulations for the thymol-Cucurbit[7]uril complex using AutoDock Vina were performed for benchmarking. Kick-MEP was validated across different molecular systems, demonstrating its effectiveness in identifying the lowest energy structures, including global minima and relevant local minima, while maintaining a low computational cost.
KW - Atomic clusters
KW - Density functional theory
KW - Electrostatic potential
KW - Molecular clusters
KW - Potential energy surface
UR - http://www.scopus.com/inward/record.url?scp=85205801165&partnerID=8YFLogxK
U2 - 10.1007/s00894-024-06155-0
DO - 10.1007/s00894-024-06155-0
M3 - Article
C2 - 39377846
AN - SCOPUS:85205801165
SN - 1610-2940
VL - 30
JO - Journal of Molecular Modeling
JF - Journal of Molecular Modeling
IS - 11
M1 - 369
ER -