Date of Award

Winter 2015

Project Type


Program or Major

Chemical Engineering

Degree Name

Master of Science

First Advisor

Harish Vashisth

Second Advisor

Kang Wu

Third Advisor

Russell T Carr


Molecular simulations of protein-nucleic acid complexes, as well as the HIV-1 Trans Activation Response Element (TAR) RNA molecule, were conducted. First, three different molecular dynamics techniques were studied on the molecule HIV-1 TAR RNA. The three techniques studied were classical molecular dynamics, steered molecular dynamics (SMD), and metadynamics. The classic molecular dynamics simulations were used to equilibrate the HIV-1 TAR RNA system, as well as every other system studied in this thesis. The SMD technique was used in order to observe the breaking force of the nucleotide interactions within TAR. This breaking force averaged to about 100pN. The metadynamics technique was used in order to accelerate the folding of HIV-1 TAR RNA from an unfolded state to its native state. With the use of root mean square deviation (RMSD) and radius of gyration (RGYR) as collective variables (CVs) we were not able to successfully fold HIV-1 TAR RNA xiv from an unfolded state to it’s native state, however, we did obtain four unique conformations of TAR that were within 1kcal/mol of the native state in free energy. Next, the classification of interaction strength between nine diverse nucleic acidprotein complexes was studied using the SMD technique. The nine chosen complexes vary in size (800-6000 atoms) as well as in the type of RNA binding protein (RBP) bound to RNA. In these simulations the RNA molecule in each system is partially fixed and the protein atoms in the binding interface are pulled at a constant velocity. Force data is obtained for each of the nine systems and the maximum force required to separate the molecules is compared using two different variables, percent composition of charged amino acid residues in the binding interface (percent composition) and buried surface area (BSA). We also look at the van der Waals and electrostatic interactions of each system over their respective trajectories. It was found that an increase in BSA often resulted in a higher value of the maximum force. The percent composition did not correlate well with the maximum force, however it is shown that the arginine rich motif (1ETG) system surprisingly had a relatively high maximum force value for such a small BSA and system size. Lastly, the binding affinity of an arginine residue bound to RNA and an adenine monophosphate (AMP) molecule bound to RNA is determined using the well-tempered metadynamics technique. Binding affinity is an important aspect to drug targeting. An effective characterization of a molecules binding affinity is the free energy of binding. Finding a way to calculate this value using molecular dynamics simulations could save much time in the drug development process. We apply well-tempered metadynamics to two small molecule systems that resemble drug-like molecular systems in order to xv determine the binding free energy of these systems. The aim here was to first test the technique on these two example systems such that the same process could be repeated for any system involving the binding of drug molecules to proteins or nucleic acids. Using welltempered metadynamics with a center-of-mass distance CV we were able to successfully determine the binding free energy of the two model systems.