메뉴 건너뛰기




Volumn 14, Issue 1, 1996, Pages 117-135

Hydrogen bonding and stacking of dna bases: A review of quantum-chemical ab initio studies

Author keywords

[No Author keywords available]

Indexed keywords

DNA BASE;

EID: 0029788346     PISSN: 07391102     EISSN: 15380254     Source Type: Journal    
DOI: 10.1080/07391102.1996.10508935     Document Type: Article
Times cited : (227)

References (122)
  • 30
    • 85023900473 scopus 로고    scopus 로고
    • order to speed up the numerical evaluation of integrals, the atomic orbitals are approximated by fixed combinations of Gauss-type functions
    • In order to speed up the numerical evaluation of integrals, the atomic orbitals are approximated by fixed combinations of Gauss-type functions
  • 55
    • 85023926975 scopus 로고    scopus 로고
    • The factors mentioned are theoretical factors, which may be somewhat reduced by an effective programming and computer implementation. This is especially well done for the HF approximation, where the actual “computer requirements” factor can be below 3
    • The factors mentioned are theoretical factors, which may be somewhat reduced by an effective programming and computer implementation. This is especially well done for the HF approximation, where the actual “computer requirements” factor can be below 3
  • 69
    • 0028546324 scopus 로고
    • CVFF - Discover, Version 2.9.5&94.0, May 1994, Biosym Technoloogies, San Diego, Ca.; CFF95 - A.T. Hagler and C.S, Ewig,; CFF95 force field has not yet been released
    • 1994. Comput. Phys. Commun., 84:131 CVFF - Discover, Version 2.9.5&94.0, May 1994, Biosym Technoloogies, San Diego, Ca.; CFF95 - A.T. Hagler and C.S, Ewig,; CFF95 force field has not yet been released
    • (1994) Comput. Phys. Commun. , vol.84 , pp. 131
  • 89
    • 36449004618 scopus 로고
    • This estimate is based on a comparison of HF/6–31G**-optimized and MP2/6–31G**-optimized geometries of cytosine dimer; the length of hydrogen bonds of the MP2-optimized structure has been corrected for the basis set superposition error (17). Let us note that a comparison with crystal data is not straightforward, because the H-bond lengths may be significantly influenced by the crystal field effects - for more details see S. Suhai
    • 1995. J. Chem. Phys., 103:7030 This estimate is based on a comparison of HF/6–31G**-optimized and MP2/6–31G**-optimized geometries of cytosine dimer; the length of hydrogen bonds of the MP2-optimized structure has been corrected for the basis set superposition error (17). Let us note that a comparison with crystal data is not straightforward, because the H-bond lengths may be significantly influenced by the crystal field effects - for more details see S. Suhai
    • (1995) J. Chem. Phys. , vol.103 , pp. 7030
  • 90
    • 85023932383 scopus 로고    scopus 로고
    • 12,. terms) while the second term includes the change of coulombic energy when passing from HF to MP2 level. Dipole and quadrupole moments are mostly overestimated at HF level by about 10–20%
    • 12,. terms) while the second term includes the change of coulombic energy when passing from HF to MP2 level. Dipole and quadrupole moments are mostly overestimated at HF level by about 10–20%
  • 91
    • 85023907851 scopus 로고    scopus 로고
    • −1 - 36). The agreement between theory and experiment for the medium-sized basis set is due a compensation of errors: size of basis set on one hand and neglect of higher correlation energy contributions on the other hand. It should be mentioned that the experimental value referred above has been reevaluated. This indicates that outcomes of experimental studies should be taken with care, including the DNA base pairs experiments. For a long time, the only data available were gas phase interaction enthalpies from mass field spectroscopy by Yanson et al. (93). Their reliability could be questioned because determination of enthalpy from the slope of the van't Hoff curve may be ambiguous. However, they are in a good agreement with theoretical results, including MP2 studies (17). Recently, Dey et al. (94) reported gas phase pseudoasociation constants for DNA base pairs. We tried to reproduce their results using the calculated interaction enthalpies and the entropy contribution obtained within the rigid rotor and harmonic oscillator - ideal gas approximations (P. Hobza and J. Šponer, submitted to). The difference between experimental and theoretical values was very large. Taking into account the quality of the theoretical MP2 procedure used the experimental values mentioned are suspicious. In addition it means that the two available gas phase experimental studies on DNA base pairs are mutually inconsistent. We assume that some assumptions given in ref. 94 (reproducibility of the desorption process and/or the thermodynamic equilibrium in the beam expansion) were not fulfilled
    • −1 - 36). The agreement between theory and experiment for the medium-sized basis set is due a compensation of errors:size of basis set on one hand and neglect of higher correlation energy contributions on the other hand. It should be mentioned that the experimental value referred above has been reevaluated. This indicates that outcomes of experimental studies should be taken with care, including the DNA base pairs experiments. For a long time, the only data available were gas phase interaction enthalpies from mass field spectroscopy by Yanson et al. (93). Their reliability could be questioned because determination of enthalpy from the slope of the van't Hoff curve may be ambiguous. However, they are in a good agreement with theoretical results, including MP2 studies (17). Recently, Dey et al. (94) reported gas phase pseudoasociation constants for DNA base pairs. We tried to reproduce their results using the calculated interaction enthalpies and the entropy contribution obtained within the rigid rotor and harmonic oscillator - ideal gas approximations (P. Hobza and J. Šponer, submitted to). The difference between experimental and theoretical values was very large. Taking into account the quality of the theoretical MP2 procedure used the experimental values mentioned are suspicious. In addition it means that the two available gas phase experimental studies on DNA base pairs are mutually inconsistent. We assume that some assumptions given in ref. 94 (reproducibility of the desorption process and/or the thermodynamic equilibrium in the beam expansion) were not fulfilled
    • Chem. Phys. Lett.
  • 114
    • 85023955988 scopus 로고    scopus 로고
    • Due to the importance of dispersion energy, we do not know whether the stacking energies are overestimated or underestimated. The use of large basis set would increase the stabilization energy, however, higher-order correlation contributions could significantly shift the intermolecular correlation interaction energy in the opposite direction. The balance between these two trends is not known, because it is different for different systems
    • Due to the importance of dispersion energy, we do not know whether the stacking energies are overestimated or underestimated. The use of large basis set would increase the stabilization energy, however, higher-order correlation contributions could significantly shift the intermolecular correlation interaction energy in the opposite direction. The balance between these two trends is not known, because it is different for different systems
  • 120
    • 85023902729 scopus 로고    scopus 로고
    • These data were obtained at the HF/4–31G level with dispersion energy evaluated by the second-order sum-of-state perturbation method (W. A. Lathan, G. R. Pack and K. Morokuma, (1975)), and represented benchmark values for nearly a decade. Our MP2/6–31G* calculations qualitatively improve these data by addresing two following items. I) The method applied by Aida can evaluate the dispersion attraction, but it does not take into consideration the intrasystem correlation interaction energy. The intrasystem correlation interaction energy reduces the electrostatic (dipole - dipole) interaction compared with the HF level II) To evaluate the interaction energy properly, the basis set of atomic Orbitals must be flexible enough to cover the space between the interacting monomers; i.e., it must contain diffuse and polarization functions. 4–31G basis set does not fulfil any of these conditions, while our basis set contains one set of d-polarization functions especially modified to evaluate the dispersion contributions (diffuse polarization functions with an exponent of 0.25)
    • J. Am. Chem. Soc., 976624 These data were obtained at the HF/4–31G level with dispersion energy evaluated by the second-order sum-of-state perturbation method (W. A. Lathan, G. R. Pack and K. Morokuma, (1975)), and represented benchmark values for nearly a decade. Our MP2/6–31G* calculations qualitatively improve these data by addresing two following items. I) The method applied by Aida can evaluate the dispersion attraction, but it does not take into consideration the intrasystem correlation interaction energy. The intrasystem correlation interaction energy reduces the electrostatic (dipole - dipole) interaction compared with the HF level II) To evaluate the interaction energy properly, the basis set of atomic Orbitals must be flexible enough to cover the space between the interacting monomers; i.e., it must contain diffuse and polarization functions. 4–31G basis set does not fulfil any of these conditions, while our basis set contains one set of d-polarization functions especially modified to evaluate the dispersion contributions (diffuse polarization functions with an exponent of 0.25)
    • J. Am. Chem. Soc. , vol.97 , pp. 6624


* 이 정보는 Elsevier사의 SCOPUS DB에서 KISTI가 분석하여 추출한 것입니다.