The use of time-averaged 3JH,H restrained molecular dynamics (tar-MD) simulations for the conformational analysis of five membered ring systems: methodology and applications

  1. Hendrickx, P.M.S. 4
  2. Corzana, F. 1
  3. Depraetere, S. 3
  4. Tourwe, D.A. 2
  5. Augustyns, K. 3
  6. Martins, J.C. 4
  1. 1 Universidad de La Rioja
    info

    Universidad de La Rioja

    Logroño, España

    ROR https://ror.org/0553yr311

  2. 2 Vrije Universiteit Brussel
    info

    Vrije Universiteit Brussel

    Bruselas, Bélgica

    ROR https://ror.org/006e5kg04

  3. 3 University of Antwerp
    info

    University of Antwerp

    Amberes, Bélgica

    ROR https://ror.org/008x57b05

  4. 4 Ghent University
    info

    Ghent University

    Gante, Bélgica

    ROR https://ror.org/00cv9y106

Revista:
Journal of Computational Chemistry

ISSN: 0192-8651

Año de publicación: 2010

Volumen: 31

Número: 3

Páginas: 561-572

Tipo: Artículo

DOI: 10.1002/JCC.21345 SCOPUS: 2-s2.0-75649102865 WoS: WOS:000273664000010 GOOGLE SCHOLAR

Otras publicaciones en: Journal of Computational Chemistry

Resumen

Because of its presence in many molecules of biological relevance, the conformational analysis of five- membered rings using 3JHH scalar coupling data from NMR is a topic of considerable interest. Typically, conformational analysis involves the use of a well-established mathematical procedure, originally developed by de Leeuw et al., that fits two rigid conformations to the available experimental data. This so-called pseudorotation analysis approach is not without problems, however, as chemically unrealistic conformations are sometimes generated from the data. Here, we present our investigations in the use of time-averaged restrained molecular dynamics simulations as a generic tool to determine the conformations that agree with experimental 3JHH scalar coupling data. For this purpose, a set of six ribose-based molecules has been used as model compounds. The influence of several modeling parameters is assessed and optimized values are proposed. The results obtained with the tar-MD approach are compared to those obtained from the two conformer fitting procedure. Interpretation of the latter is facilitated by the introduction of a fitting error analysis that allows mapping the solution space of the fitting procedure. The relative merits of both methods and the advantages that result from the use of a force field and a time-averaged restraint potential for the experimental data are discussed. When combined, both techniques allow an enhanced understanding of the molecules' conformational behavior and prevent possible overinterpretation. In view of the very reasonable computational burden of a tar-MD simulation for the systems investigated here, the approach should be generally applicable. © 2009 Wiley Periodicals, Inc.