Detailseite
Projekt Druckansicht

Multiresolution numerical methods for the integral equation theory of solute-solvent interactions in molecular liquids

Fachliche Zuordnung Theoretische Chemie: Elektronenstruktur, Dynamik, Simulation
Förderung Förderung von 2010 bis 2014
Projektkennung Deutsche Forschungsgemeinschaft (DFG) - Projektnummer 181654620
 
Erstellungsjahr 2015

Zusammenfassung der Projektergebnisse

We have developed the fast and accurate multi-grid numerical methods for solving large systems of integral equations of the molecular integral equations theory. The use of the multi-resolution/multi-grid approaches allows us to accelerate the calculations by an order of magnitude compared to the methods traditionally used in the field. We have applied the methods for the most popular molecular theories of liquids - the RISM and the 3DRISM. These models calculate the structural and thermodynamic properties of molecules solvated in liquid media, including the Solvation Free Energy (SFE). SFE is the fundamental thermodynamic quantity in solution chemistry and chemical physics of liquid state. It determines solubility of a molecular compound in a given solvent and its concentration distribution (partition) between different compartments (e.g. gas-water or oil-water partition coefficients). The standard SFE prediction methods are either computationally too expensive (e.g. a calculation of SFE for one molecule by Molecular Dynamics simulations would require days of the computing time) or too inaccurate for the practical application. In combination with the based on RISM semi-empirical theories (developed in the course of the project), the new numerical methods led us to development of a new fast computational tool that accurately predicts the SFE of a wide variety of organic compounds in water. The computational costs of one SFE calculation by the new method is of the order of seconds to minutes only, so it can be used to high-throughput screening of the large databases of drug candidate molecules for desirable physical-chemical characteristics (e.g. solubility, lipophilicity etc). The significant improvement of the numerical performance of RISM algorithms with the use of the multi grid methods was a positive surprise itself - we have expected the speed up factor to be around 2 to 3, but gained at least an order of magnitude. Another positive surprise was the development of the semi-empirical model for the accurate calculations of SFE by RISM (the UC model discussed above). This “byproduct” of the project has recently attracted much attention in the theoretical community (for about 40 years the RISM theory has suffered from the poor accuracy of SFE predictions).

Projektbezogene Publikationen (Auswahl)

  • 2010. Towards a universal method for calculating hydration free energies: a 3D reference interaction site model with partial molar volume correction. Journal of Physics: Condensed Matter 22, 492101
    Palmer, D.S., Frolov, A.I., Ratkova, E.L., Fedorov, M.V.
  • 2011. Hydration Thermodynamics Using the Reference Interaction Site Model: Speed or Accuracy? Journal of Physical Chemistry B 115, 6011–6022
    Frolov, A.I., Ratkova, E.L., Palmer, D.S., Fedorov, M.V.
  • 2011. Multigrid Solver for the Reference Interaction Site Model of Molecular Liquids Theory. Journal of Computational Chemistry 32, 1982–1992
    Sergiievskyi, V.P., Hackbusch, W., Fedorov, M.V.
  • 2012. 3DRISM Multigrid Algorithm for Fast Solvation Free Energy Calculations. Journal of Chemical Theory and Computation, 8, 2062–2070
    Sergiievskyi, V.P., Fedorov, M.V.
    (Siehe online unter https://doi.org/10.1021/ct200815v)
  • 2012. First-Principles Calculation of the Intrinsic Aqueous Solubility of Crystalline Druglike Molecules. Journal of Chemical Theory and Computation, 8, 3322–3337
    Palmer, D.S., McDonagh, J.L., Mitchell, J.B.O., van Mourik, T., Fedorov, M.V.
    (Siehe online unter https://doi.org/10.1021/ct300345m)
  • 2013. Solvent Binding Analysis and Computational Alanine Scanning of the Bovine Chymosin–Bovine κ-Casein Complex using Molecular Integral Equation Theory. Journal of Chemical Theory and Computation, 9, 5706––5717
    Palmer, D.S., Sørensen, J., Schiøtt, B., Fedorov, M.V.
    (Siehe online unter https://doi.org/10.1021/ct400605x)
 
 

Zusatzinformationen

Textvergrößerung und Kontrastanpassung