Stochastic solutions

The applications of this stochastic picture to the trans-gauche isomerization of -alkanes and the boat-chair isomerization of cyclohexane have demonstrated the usefulness of this approach. In both cases, the transmission coefficients calculated from stochastic dynamics agreed quite well with those from the (later) molecular dynamics calculations, given that there can be an uncertainty in the correct value of the collision frequency to use in comparing with the full molecular dynamics in solution. Stochastic dynamics therefore can allow the rapid calculation of reaction dynamics over a wide range of solvent densities and/or viscosities. 

Let u be a vector valued stochastic variable with dimension D x 1 and with covariance matrix Ru of size D x D. The key idea is to linearly transform all observation vectors, u , to new variables, z = W Uy, and then solve the optimization problem (1) where we replace u, by z . We choose the transformation so that the covariance matrix of z is diagonal and (more importantly) none if its eigenvalues are too close to zero. (Loosely speaking, the eigenvalues close to zero are those that are responsible for the large variance of the OLS-solution). In order to liiid the desired transformation, a singular value decomposition of /f is performed yielding... 

Kramers solution of the barrier crossing problem [45] is discussed at length in chapter A3.8 dealing with condensed-phase reaction dynamics. As the starting point to derive its simplest version one may use the Langevin equation, a stochastic differential equation for the time evolution of a slow variable, the reaction coordinate r, subject to a rapidly statistically fluctuating force F caused by microscopic solute-solvent interactions under the influence of an external force field generated by the PES F for the reaction... 

Since the stochastic Langevin force mimics collisions among solvent molecules and the biomolecule (the solute), the characteristic vibrational frequencies of a molecule in vacuum are dampened. In particular, the low-frequency vibrational modes are overdamped, and various correlation functions are smoothed (see Case [35] for a review and further references). The magnitude of such disturbances with respect to Newtonian behavior depends on 7, as can be seen from Fig. 8 showing computed spectral densities of the protein BPTI for three 7 values. Overall, this effect can certainly alter the dynamics of a system, and it remains to study these consequences in connection with biomolecular dynamics. 

We further discuss how quantities typically measured in the experiment (such as a rate constant) can be computed with the new formalism. The computations are based on stochastic path integral formulation [6]. Two different sources for stochasticity are considered. The first (A) is randomness that is part of the mathematical modeling and is built into the differential equations of motion (e.g. the Langevin equation, or Brownian dynamics). The second (B) is the uncertainty in the approximate numerical solution of the exact equations of motion. 

The two sources of stochasticity are conceptually and computationally quite distinct. In (A) we do not know the exact equations of motion and we solve instead phenomenological equations. There is no systematic way in which we can approach the exact equations of motion. For example, rarely in the Langevin approach the friction and the random force are extracted from a microscopic model. This makes it necessary to use a rather arbitrary selection of parameters, such as the amplitude of the random force or the friction coefficient. On the other hand, the equations in (B) are based on atomic information and it is the solution that is approximate. For ejcample, to compute a trajectory we make the ad-hoc assumption of a Gaussian distribution of numerical errors. In the present article we also argue that because of practical reasons it is not possible to ignore the numerical errors, even in approach (A). 

One of the main advantages of the stochastic dynamics methods is that dramatic tirn savings can he achieved, which enables much longer stimulations to he performed. Fc example, Widmalm and Pastor performed 1 ns molecular dynamics and stochastic dynamic simulations of an ethylene glycol molecule in aqueous solution of the solute and 259 vvatc jnolecules [Widmalm and Pastor 1992]. The molecular dynamics simulation require 300 hours whereas the stochastic dynamics simulation of the solute alone required ju 24 minutes. The dramatic reduction in time for the stochastic dynamics calculation is du not only to the very much smaller number of molecules present hut also to the fact the longer time steps can often he used in stochastic dynamics simulations. 

The difficulty with Eq, (26-58) is that it is impossible to determine the velocity at every point, since an adequate turbulence model does not currently exist, The solution is to rewrite the concentration and velocity in terms of an average and stochastic quantity C = (C) -t- C Uj = (uj) + Uj, where the brackets denote the average value and the prime denotes the stochastic, or deviation variable. It is also helpful to define an eddy diffusivity Kj (with units of area/time) as... 

In finite boundary conditions the solute molecule is surrounded by a finite layer of explicit solvent. The missing bulk solvent is modeled by some form of boundary potential at the vacuum/solvent interface. A host of such potentials have been proposed, from the simple spherical half-harmonic potential, which models a hydrophobic container [22], to stochastic boundary conditions [23], which surround the finite system with shells of particles obeying simplified dynamics, and finally to the Beglov and Roux spherical solvent boundary potential [24], which approximates the exact potential of mean force due to the bulk solvent by a superposition of physically motivated tenns. 

The literature of science is replete with models. This variety enables one to make some interesting observations. Thus, for example, one rarely regards models as unique or absolute, although, through the choice of a specific one (e.g., a differential equation), unique solutions to problems may be obtained. A model is formulated to serve a specific purpose. Some models may be suitable for generalization, others may not be. These generalizations are more profitably made as extrapolations for scientific purposes, and occasionally as useful philosophical observations. A model must be flexible to absorb new information, and, hence, stochastic processes have broader and richer applicability than deterministic models. 

Takaes, Lajos, Stochastic Processes, Problems and Solutions, Methuen s Monographs on Applied Probability and Statistics, John Wiley and Sons, Inc., New York, 1960. 

Both of the numerical approaches explained above have been successful in producing realistic behaviour for lamellar thickness and growth rate as a function of supercooling. The nature of rough surface growth prevents an analytical solution as many of the growth processes are taking place simultaneously, and any approach which is not stochastic, as the Monte Carlo in Sect. 4.2.1, necessarily involves approximations, as the rate equations detailed in Sect. 4.2.2. At the expense of... 

When the film thickness is of the order of roughness heights, the effects of roughness become significant which have to be taken into account in a profound model of mixed lubrication. The difficulty is that the stochastic nature of surface roughness results in randomly fluctuating solutions that the numerical techniques in the 1970s are unable to handle. As... 

Special considerations are required in estimating paraimeters from experimental measurements when the relationship between output responses, input variables and paraimeters is given by a Monte Carlo simulation. These considerations, discussed in our first paper 1), relate to the stochastic nature of the solution and to the fact that the Monte Carlo solution is numerical rather than functional. The motivation for using Monte Carlo methods to model polymer systems stems from the fact that often the solution... 

The method for estimating parameters from Monte Carlo simulation, described in mathematical detail by Reilly and Duever (in preparation), uses a Bayesian approach to establish the posterior distribution for the parameters based on a Monte Carlo model. The numerical nature of the solution requires that the posterior distribution be handled in discretised form as an array in computer storage using the method of Reilly 2). The stochastic nature of Monte Carlo methods implies that output responses are predicted by the model with some amount of uncertainty for which the term "shimmer" as suggested by Andres (D.B. Chambers, SENES Consultants Limited, personal communication, 1985) has been adopted. The model for the uth of n experiments can be expressed by... 

We have presented applications of a parameter estimation technique based on Monte Carlo simulation to problems in polymer science involving sequence distribution data. In comparison to approaches involving analytic functions, Monte Carlo simulation often leads to a simpler solution of a model particularly when the process being modelled involves a prominent stochastic coit onent. 

In this work, therefore we aim to combine the stochastic observer to input/output prediction model so that it can be robust against the influence of noise. We employ the modified I/O data-based prediction model [3] as a linear part of Wimra" model to design the WMPC and these controllers are applied to a continuous mefihyl methacrylate (MMA) solution polymerization reactor to examine the performance of controller. 

All conventional approaches (mathematical and stochastic programming, parametric and nonparametric regression analysis) adopt as a common solution format real vectors, x and as performance criterion,... 

Kuznetsov, A. M., Stochastic and Dynamic Views of Chemical Reaction Kinetics in Solutions, Presses Polytechniques et Universitaires Romandes, Lausanne, Switzerland, 1999. Kuznetsov, A. M., and J. Ulstrup, Electron Transfer in Chemistry and Biology, Wiley, Chichester, West Sussex, England, 1999. 

Depending on whether or not stochastic features are introduced in the simulation procedure, simulation methods are sometimes classified as stochastic or deterministic. Although the second term is usually applied to methods related to the numerical solution of Newton s equations, the first term is applied to a wide variety of simulation metfiods. 

The second step is the molecular dynamics (MD) calculation that is based on the solution of the Newtonian equations of motion. An arbitrary starting conformation is chosen and the atoms in the molecule can move under the restriction of a certain force field using the thermal energy, distributed via Boltzmann functions to the atoms in the molecule in a stochastic manner. The aim is to find the conformation with minimal energy when the experimental distances and sometimes simultaneously the bond angles as derived from vicinal or direct coupling constants are used as constraints. 

The conductivity of membranes that do not contain dissolved ionophores or lipophilic ions is often affected by cracking and impurities. The value for a completely compact membrane under reproducible conditions excluding these effects varies from 10-8 to 10 10 Q 1 cm-2. The conductivity of these simple unmodified membranes is probably statistical in nature (as a result of thermal motion), due to stochastically formed pores filled with water for an instant and thus accessible for the electrolytes in the solution with which the membrane is in contact. Various active (natural or synthetic) substances... 

It is also worth noting that the stochastic optimization methods described previously are readily adapted to the inclusion of constraints. For example, in simulated annealing, if a move suggested at random takes the solution outside of the feasible region, then the algorithm can be constrained to prevent this by simply setting the probability of that move to 0. 

---