MOLECULAR SIMULATION GROUP
Prof. Rodrigo L. Silveira | Institute of Chemistry - UFRJ
MOLECULAR DYNAMICS
Classical molecular dynamics consists in numerically solving the equations of motion of a molecular system, under interaction potentials previously determined (force fields).
Electrons are not taken in account explicitly, so this method allows us to explore the dynamics of systems containing, tipically, 10⁶ atoms in timescales ranging from nanoseconds to microseconds.
We use MD to study the dynamics of enzymes, their interactions with substrates and products, and conformational changes relevant to understand their mechanisms.
QUANTUM CHEMISTRY
Quantum Chemistry methods deal explicitly with all electrons comprising the system and so require an approximate solution to the Schoroedinger equation.
QC methods are computationally expensive; hence, they can be applied only to systems containing no more than hundreds of atoms, depending on the level of calculation.
We use QC methods to unravel details of molecular events in which an explicit description of electrons and orbitals is needed, such as reactions in metalloenzymes.
QM/MM
QM/MM (quantum mechanics/molecular mechanics) methods consist in simultaneously employing classical force fields and quantum mechanics to treat the system.
The part of the system where quantum effects are relevant, such as the active site of an enzyme, is treated with a quantum chemistry method. On the other hand, the usually larger part of the system where quantum effects do not need to be explicitly accounted is treated with a classical force field.
Such partitioning allow the heavier computational effort to be concentrated in the small quantum region of the system, resulting in a dramatic decrease of the cost when compared to the full quantum description of the system. We use QMMM to study the dynamics of enzymatic reactions.
ENHANCED SAMPLING AND FREE ENERGY CALCULATIONS
Enhanced sampling techniques allow us to overcome high energy barriers within the system. Thus, we can explore rare events such as chemical reactions, substrate binding, product release, large-scale conformational changes, and others, with ameanable computational costs.
Being able to efficiently sample relevant regions of the conformational space through enhanced sampling techniques, gives us the possibility to calculate free energy profiles associated to different steps of an enzyme's mechanism.