Compensating Mass Matrix Potential for Constrained Molecular Dynamics

has been a major hurdle on the use of these compensating potentials more generally in constrained MD simulations. Rigid internal constraints are used in molecular models to speed up molecular dynamics (MD) simulations. It is well recognized that This paper analyzes the structure of the compensating statistical averages from such constrained MD simulations differ by potential and its gradient and develops substantially sima metric tensor-dependent term from similar averages computed pler expressions for them for tree topology molecular modusing conventional unconstrained MD simulations. Fixman proels. These expressions are used to derive computational posed augmenting the standard potential with a compensating term algorithms for use in constrained MD simulations. The which depends on the metric tensor to nullify the effects of this algorithms are straightforward extensions of the recently bias term. However, in the absence of tractable algorithms to compute this compensating tensor potential and its gradient its use has proposed spatial operators based O(N ) algorithm for conbeen impractical. This paper derives a new algorithm for computing strained MD simulations [2]. Indeed, the compensating the compensating potential, as well as its gradient for tree topology potential and its gradient are readily computable from molecular systems. The algorithm is quite straightforward and is the articulated body inertia quantities available from this an extension of the spatial operators based O(N ) algorithm that O(N ) algorithm. has been recently proposed for constrained dynamics. Indeed, the compensating potential is closely related and computed from the 2. ENSEMBLE AVERAGES IN CONSTRAINED articulated body inertia quantities available from this O(N ) DYNAMICS SIMULATIONS algorithm. Q 1997 Academic Press