Markov State Model Analysis of Haloperidol Binding to the D₃ Dopamine Receptor

Abstract

We have developed Markov state models (MSMs) and hidden Markov models (HMMs) that describe the binding of haloperidol to the D 3 dopamine receptor. Haloperidol is an antipsychotic drug that binds with nanomolar affinity to the D 3 dopamine receptor, where it functions as an inverse agonist. The models were constructed using an adaptive sampling approach from 519 individual molecular dynamics simulations totaling 122 μs of simulated time and encompass the entire drug binding process. They reveal short-lived metastable bound states and two distinct long-lived bound conformations that cannot be separated in affinity using our current methodology. This work extends the use of MSMs and HMMs to study ligand binding, which thus far has been limited to simpler systems.

Modeling Ensembles of Enzyme Reaction Pathways with Hi-MSM Reveals the Importance of Accounting for Pathway Diversity

“…Time-lagged independent component analysis (tICA) was used to accomplish this task by determining the combinations of degrees of freedom, or collective variables, that maximize the autocorrelation function, implying slow degrees of freedom. tICA has been shown to be a better choice for dimension reduction for Markov state modeling than principal component analysis (PCA), which utilizes the collective variables which maximize the variance of the data. − Active-site residues were determined to be any residue whose heavy atom falls within 4.5 Å of the heavy atoms in the rings of the substrates where the hydride transfer occurs in any frame over the course of a 50 ns simulation. A visual representation of these residues can be seen in Figure a.…”

Nonparametric Analysis of Nonequilibrium Simulations

“…For systems with a relatively simple, small configuration space, selection of initial configurations to start many short simulations, as well as the seed RC, can be done analytically, as it was done for the model system considered here. Such systems may include practically important cases such as, e.g., studies of dynamics of a ligand binding/unbinding to/from a protein ,, or diffusion of a small molecule/ion through an ion channel pore.…”

“…One general strategy in overcoming the sampling problem in biomolecular simulations consists of simulating a very large ensemble of short trajectories rather than a singe long trajectory. This strategy allows seamless parallelization and is a promising approach toward simulations employing exascale or cloud computing. , Adaptive sampling approaches can be considered as an extension of this strategy, where one, for example, improves sampling in less sampled parts of configuration space or parts that produce the largest error or controls the exploration/exploitation balance. − The swarms of trajectories , are another successful variation of this idea.…”

Non-Parametric Analysis of Non-Equilibrium Simulations

“…For systems with relatively simple, small configuration space, selection of initial configurations to start many short simulations as well as the seed RC can be done analytically, as it was done for the model system considered here. Such systems may include practically important cases such as, e.g., studies of dynamics of a ligand binding/unbinding to/from a protein, 8,9 or diffusion of a small molecule/ion through an ion channel pore.…”

“…1,2 Adaptive sampling approaches can be considered as an extension of this strategy, where one, for example, improves sampling in less sampled parts of configuration space, or parts that produce largest error or controls the exploration/exploitation balance. [3][4][5][6][7][8][9] The swarms of trajectories 10,11 is another successful variation of this idea.…”