Access by Xinjiang University

Molecular dynamics sampling scheme realizing multiple distributions

¹National Institute of Advanced Industrial Science and Technology, 2-41-6, Aomi, Koto-ku, Tokyo 135-0064, Japan
²Institute for Protein Research, Osaka University, 3-2 Yamadaoka, Suita, Osaka 565-0871, Japan

Phys. Rev. E 71, 046708 – Published 25 April, 2005

DOI: https://doi.org/10.1103/PhysRevE.71.046708

Abstract

We present a molecular-dynamics sampling scheme in which any summation of multiple arbitrarily given distributions can be realized deterministically by a certain theoretical guide. Our scheme thus provides a static view of the composition of the distributions, rather than a dynamic view in which some parameter is switched at a certain time in the simulation process. The proposed method induces the orbit to jump automatically among different phase space regions, without the use of any artificial timing parameter. In addition, the proposed method does not require an assumption that the density of states is smooth. We applied it to multiple Tsallis distributions and established a suitable series of parameter values for which the sum of the distributions allows broad sampling. Numerical simulations applied to fundamental models with multi-Tsallis distributions showed efficient sampling, characterized by an energy trajectory that was totally different from that associated with each single distribution.

Article Text

References (60)

F. Faupel, W. Frank, M.-P. Macht, H. Mehrer, V. Naundorf, K. Räzke, H. R. Schober, S. K. Sharma, and H. Teichler, Rev. Mod. Phys. 75, 237 (2003).
S. Sastry, Nature (London) 409, 164 (2001).
N. A. Alves and U. H. E. Hansmann, Phys. Rev. Lett. 84, 1836 (2000).
W. G. Hoover and B. L. Holian, Phys. Lett. A 211, 253 (1996), and the references therein.
W. G. Hoover, Computational Statistical Mechanics (Elsevier, New York, 1991).
Y. Sugita and Y. Okamoto, Chem. Phys. Lett. 314, 141 (1999).
Y. Sugita, A. Kitao, and Y. Okamoto, J. Chem. Phys. 113, 6042 (2000).
U. H. E. Hansmann, Y. Okamoto, and F. Eisenmenger, Chem. Phys. Lett. 259, 321 (1996).
N. Nakajima, H. Nakamura, and A. Kidera, J. Phys. Chem. B 101, 817 (1997).
S. Jang, Y. Pak, and S. Shin, J. Chem. Phys. 116, 4782 (2002).
J. G. Kim, Y. Fukunishi, A. Kidera, and H. Nakamura, Phys. Rev. E 68, 021110 (2003).
A. Mitsutake and Y. Okamoto, J. Chem. Phys. 121, 2491 (2004), and the references therein.
R. Zhou, B. J. Berne, and R. Germain, Proc. Natl. Acad. Sci. U.S.A. 98, 14931 (2001).
A. E. Garcia and J. N. Onuchic, Proc. Natl. Acad. Sci. U.S.A. 100, 13898 (2003).
R. Yamamoto and W. Kob, Phys. Rev. E 61, 5473 (2000).
K. Ikeda, O. V. Galzitskaya, H. Nakamura, and J. Higo, J. Comput. Chem. 24, 310 (2003).
N. Nakajima, J. Higo, A. Kidera, and H. Nakamura, Chem. Phys. Lett. 278, 297 (1997).
Y. S. Watanabe, Y. Fukunishi, and H. Nakamura, J. Mol. Graphics Modell. 23, 59 (2004).
N. Kamiya, J. Higo, and H. Nakamura, Protein Sci. 11, 2297 (2002).
S. Ono, N. Nakajima, J. Higo, and H. Nakamura, J. Comput. Chem. 21, 748 (2000).
C. Tsallis, J. Stat. Phys. 52, 479 (1988).
C. Tsallis, Braz. J. Phys. 29, 1 (1999), for an updated bibliography, see http://tsallis.cat.cbpf.br/biblio.htm
I. Andricioaei and J. E. Straub, J. Chem. Phys. 107, 9117 (1997).
U. H. E. Hansmann, F. Eisenmenger, and Y. Okamoto, Chem. Phys. Lett. 297, 374 (1998).
Y. Pak and S. Wang, J. Chem. Phys. 111, 4359 (1999).
A. R. Plastino and C. Anteneodo, Ann. Phys. (N.Y.) 255, 250 (1997).
I. Fukuda and H. Nakamura, Phys. Rev. E 65, 026105 (2002).
J. S. Andrade, Jr., M. P. Almeida, A. A. Moreira, and G. A. Farias, Phys. Rev. E 65, 036121 (2002).
I. Fukuda and H. Nakamura, Chem. Phys. Lett. 382, 367 (2003).
I. Fukuda and H. Nakamura, J. Phys. Chem. B 108, 4162 (2004).
A. M. Ferrenberg and R. H. Swendsen, Phys. Rev. Lett. 61, 2635 (1988).
U. H. E. Hansmann, Phys. Rev. E 56, 6200 (1997).
T. Terada, Y. Matsuo, and A. Kidera, J. Chem. Phys. 118, 4306 (2003).
I. Fukuda and H. Nakamura, in Slow Dynamics in Complex Systems: Third International Symposium on Slow Dynamics in Complex Systems, edited by Michio Tokuyama and Irwin Oppenheim, AIP Conf. Proc. No. 708 (AIP, Woodbury, 2004), p. 356.
S. Jang, S. Shin, and Y. Pak, Phys. Rev. Lett. 91, 058305 (2003).
S. Nosé, J. Chem. Phys. 81, 511 (1984).
W. G. Hoover, Phys. Rev. A 31, 1695 (1985).
We can also say “ergodicity for $ρ$ [Eq. (10)],” since the ergodicity is an invariant property with respect to a constant multiplication $\hat{s} : ρ \mapsto \tilde{ρ} \equiv s ρ (0 < s < \infty)$ . So Eq. (15) is concluded directly rather than through Eq. (14). Only change due to $\hat{s}$ is for the invariant measure, as $P \mapsto \tilde{P} \equiv s P$ ; note that the one-parameter group of automorphism for measure space $(Γ, L, P)$ also becomes that for $(Γ, L, \tilde{P})$ , consistent with the invariance of Eq. (1) with respect to $\hat{s}$ , and so the invariance is inherited in Eqs. (8) and (12) [or Eq. (9)].
R. W. Zwanzig, J. Chem. Phys. 22, 1420 (1954).
J. G. Kirkwood, J. Chem. Phys. 3, 300 (1935).
N. D. Lu, J. K. Singh, and D. A. Kofke, J. Chem. Phys. 118, 2977 (2003), and references therein.
For a more accurate and general description about the limit for $∣ I ∣ \to 0$ , refer to, e.g., G. B. Folland, Real Analysis (Wiley, New York, 1984), p. 93.
In the present case, from the standpoint of the escort probabilities [44], the form of $b_{a} \equiv q_{a} ∕ (1 - q_{a})$ may not be so useful, since the escort probabilities will lead to a form such as ${(\sum_{a = 1}^{M} ρ^{a})}^{q}$ and not $\sum_{a = 1}^{M} {(ρ^{a})}^{q_{a}}$ ; but the form $b_{a} \equiv q_{a} ∕ (1 - q_{a})$ is useful for expediting the finiteness of each partition function, retaining the difference with the BG distribution [30].
C. Tsallis, R. S. Mendes, and A. R. Plastino, Physica A 261, 534 (1998).
A. R. Plastino and A. Plastino, Phys. Lett. A 193, 140 (1994).
D. Kusnezov, A. Bulgac, and W. Bauer, Ann. Phys. (N.Y.) 204, 155 (1990).
I. L’Heureux and I. Hamilton, Phys. Rev. E 47, 1411 (1993).
S. Nosé, Prog. Theor. Phys. Suppl. 103, 1 (1991).
A. C. Brańka, M. Kowalik, and K. W. Wojciechowski, J. Chem. Phys. 119, 1929 (2003).
I. Andricioaei, A. R. Dinner, and M. Karplus, J. Chem. Phys. 118, 1074 (2003).
B. C. Dian, A. Longarte, and T. S. Zwier, Science 296, 2369 (2002).
D. J. Tobias, S. F. Sneddon, and C. L. Brooks III, J. Mol. Biol. 216, 783 (1990).
W. D. Cornell, P. Cieplak, C. I. Bayly, I. R. Gould, K. M. Merz, Jr., D. M. Ferguson, D. C. Spellmeyer, T. Fox, J. W. Caldwell, and P. A. Kollman, J. Am. Chem. Soc. 117 5179 (1995).
P. Kollman, R. Dixon, W. Cornell, T. Fox, C. Chipot, and A. Pohorille, in Computer Simulation of Biomolecular Systems, edited by W. F. van Gunsteren, P. K. Weiner, and A. J. Wilkinson (Kluwer, Netherlands, 1997), Vol. 3, pp. 83–96.
K. Morikami, T. Nakai, A. Kidera, M. Saito, and H. Nakamura, J. Comput. Chem. 16, 243 (1992).
Y. Fukunishi, Y. Mikami, and H. Nakamura, J. Phys. Chem. B 107, 13201 (2003).
U. H. E. Hansmann and Y. Okamoto, Phys. Rev. E 56, 2228 (1997).
This is intuitively obvious. Also in a mathematically rigorous sense, using elementary facts, the results can be obtained from the assumption that a function $f$ on a closed interval $J \subset R$ is concave, i.e., $f (λ x + (1 - λ) y) ⩾ λ f (x) + (1 - λ) f (y)$ for all $x, y ∊ J$ and all $λ ∊ (0, 1)$ : If $f (E_{1}) = f (E_{2})$ holds for inner points $E_{1}, E_{2}, (E_{1} < E_{2}) ∊ \overset{̊}{J}$ , then a maximum point $E_{\max}$ exists in $[E_{1}, E_{2}]$ , viz., $f (E_{\max}) = \max f (J)$ . The assumption for smoothness of $f$ is not needed.
N. D. Lu and D. A. Kofke, J. Chem. Phys. 114, 7303 (2001).
A. Mitsutake, Y. Sugita, and Y. Okamoto, J. Chem. Phys. 118, 6664 (2003).

Physical Review E

Export citation

Export citation

Molecular dynamics sampling scheme realizing multiple distributions

Abstract

Article Text

References (60)

Outline

Information

Filter

Filter

Article Lookup

Enter a citation