During this fiscal year we devoted major effort to work aimed at applications of molecular dynamics and quantum mechanics/molecular mechanics simulations required to help support the computational chemistry and molecular modeling needs of NIEHS scientists. Some projects involved creation of solution structures of peptides and proteins using state-of-the-art molecular dynamics simulations and the others involved a careful look at the reactive dynamics at or near the active site of the biological systems of interest. Several docking studies and energy characterization studies are highlights of our efforts. Most computational chemistry and molecular modeling tools that have been utilized in the present research efforts are either developed by us or modified by us. Almost all tools used in the analysis of molecular dynamics trajectories required to obtain predicted solution structures and in the energy decomposition schemes of QMMM calculations are also written by us. The current list of projects includes (but not limited to) mutational studies of Tristetraproline (a protein involved in RNA degradation) that affects RNA binding; mutational studies of Aprataxin, dynamics of HIV reverse transcriptase and its constituents (P51 and P66) along with some critical mutations that effect the function, construction of a solution structures for human constitutively active receptor (hCAR) and running long molecular dynamics of a related protein PPARgamma, characterization of Pinobarbital binding to EGF receptor, modeling of DNA polymerase activity with the inclusion of some ribonucleotides in the DNA sequence at a classical level, ribonucleotide insertion in DNA during polymerization catalysis using QMMM methods, interaction predictions of various CNOT proteins, structure and energy characterization of collagen glycopeptides, a transcription factor MRG15 and HNF4-alpha interactions with MRFAP1, quantum mechanical characterization of some flame retardant molecules. In addition, as a precautionary measure to carry out our functions under constraints of budgetary restrictions, we have continued to explore the idea of setting up computer servers based on low cost, off-the-shelf components and GPUs to run MD simulations that require heavy utilization of multiple processors to tackle systems with millions of atoms and to complete QMMM calculations that demand access to a large sum of memory at a given instance.
Showing the most recent 10 out of 35 publications
