Recent COMP News‎ > ‎

The Chemical Computing Group Excellence Award for Graduate Students Winners for Indianapolis (Fall 2013)

posted Apr 24, 2013, 5:48 AM by Emilio Xavier Esposito
The COMP Division is excited to announce the Chemical Computing Group Excellence Award for Graduate Students winners for the Indianapolis ACS meeting (fall 2013). Please visit the COMP award winners and the other excellent COMP posters at the COMP Poster Session on Tuesday, September 10, 2013 from 6pm to 8pm at a location to be determined. 

Theoretical investigation of ion diffusion pathways in II-VI semiconductor nanocrystals
Joseph W May, Daniel R Gamelin and Xiaosong Li; Department of Chemistry, University of Washington
Continued advances in the development of smaller, faster computer processors; cheaper, more efficient solar energy conversion; and environmentally conscious, high-capacity energy storage relies upon the design of new materials. Semiconducting nanocrystals such as II-VI (ZnO, CdSe, etc.) quantum dots (QDs) have shown promising application towards each of these endeavors. Electrically charged II-VI semiconductor nanocrystals have been successfully used in activating room-temperature magnetic ordering and as chemical reduction catalysts. It was hypothesized that the charge-compensating cations in an electrolytic solution, which serve to stabilize the conduction band electrons, are either absorbed to the QD surface or diffused into the QD lattice. This latter scenario happens to share the same mechanism underlying modern day lithium-ion batteries that rely on the intercalation (diffusion) of Li+ ions into the anode and cathode materials. This hypothesis also holds the greatest promise for the development of new materials for energy storage, that is, semiconducting nanocrystal based lithium-ion batteries. A solid understanding of the ion diffusion processes in QDs lends significant insight into the design of functional materials for reusable energy storage. In this work, computational methods such as electronic structure theory and molecular dynamics are used to examine the energetics, thermodynamics, and kinetics of charge compensating cations (H+ and Li+) in II-VI QDs. Ion diffusion pathways in II-VI semiconductor QDs are studied in an effort to understand the physical underpinnings (kinetics and thermodynamics) of ion diffusion pathways in II-VI semiconductor colloidal nanocrystals. More specifically, the kinetics and thermodynamics conditions for proton and lithium ions to diffuse in/out (charging/discharging) of the semiconductor nanocrystals are investigated.

Acceleration of ab initio quantum chemistry calculation on GPUs
Yipu Miao and Kenneth M Merz; Department of Chemistry, University of Florida
Quantum theory has been utilized in many roles, including interpreting chemical phenomena and predicting new molecular species with novel functions.For computational chemistry, the most widely used types of electronic structure calculation are Hatree-Fock (HF) and density functional theory (DFT). Typically, for these methods, Electron Repulsion Integral (ERI) calculations, together with basic linear algebraic operations dominate the computational time. We developed a series of mapping strategies to avoid global memory access latency and thread divergence. Our benchmark studies shows that GPU can speedup ERI calculation by 100 times over a traditional GPU. This speedup is achieved by optimizing the Fock formation scheme by atomic-operations in order to avoid data transfer, a well-known GPU architectural bottleneck.

Multi-receptor high-throughput virtual docking on supercomputers with VinaMPI
Sally R Ellingson,1,3 Jerome Baudry2,3 and Jeremy C Smith2,3; 1. Genome Science & Technology, University of Tennessee, Knoxville; 2. Biochemistry & Cellular & Molecular Biology, University of Tennessee, Knoxville; 3. Oak Ridge National Laboratory, Center for Molecular Biophysics, Oak Ridge, Tennessee
Virtual docking is a computational process that aims at predicting the bound conformation of a protein-ligand complex and how well it binds through a scoring algorithm. The scoring functions commonly used in docking applications use significant approximations to rapidly estimate protein-ligand binding affinities and the resulting computational efficiency make these applications useful for virtual high-throughput screens in which millions of molecules can be tested quickly. Most of the screening applications developed to date focus on docking a library of drug-like molecules into one protein target. However, inverse techniques of docking libraries of chemical compounds into a library of proteins are of significant interest, as these permit the investigation of many conformational states of a single protein thus increasing the chemical diversity of drug candidates, and of the effects of a single target compound against a range of different proteins permitting the exploration of toxicity/side-effects of the drug and polypharmacology capabilities.

VinaMPI, a massively parallel Message Passing Interface (MPI) and muti-threaded version of the virtual docking program Autodock Vina has been developed. MPI is used to distribute tasks while multi-threading is used to speed-up individual docking tasks. VinaMPI uses a distribution scheme in which tasks are evenly distributed to the workers based on the complexity of each task, as defined by the number of rotatable bonds in each chemical compound investigated. VinaMPI efficiently handles multiple proteins in a ligand screen, allowing for high-throughput inverse docking that presents new opportunities for improving the efficiency of the drug discovery pipeline. VinaMPI successfully ran on 84,672 cores with a continual decrease in job completion time with increasing core count. The ratio of the number of tasks in a screening to the number of workers should be at least around 100 in order to have a good load balance and an optimal job completion time.

Unlocking the binding and reaction mechanism of hydroxyurea as a biological nitric oxide donor
Sai Lakshmana Vankayala, Henry Lee Woodcock, and Jacqueline C Hargis; Department of Chemistry, University of South Florida
Hydroxyurea is the only FDA approved treatment of sickle cell disease. It is believed the primary mechanism of action is associated with the pharmacological elevation of nitric oxide in the blood, however, the exact details of this mechanism is still unclear. HU interacts with oxy and deoxyHb resulting in slow NO production rates. However, this did not correlate with the observed increase of NO concentrations in patients undergoing HU therapy. The discrepancy can be attributed to the interaction of HU competing with other heme based enzymes such as catalase and peroxidases. In the current work, we investigate the atomic level details of this process using a combination of flexible-ligand / flexible-receptor virtual screening (i.e. induced fit docking, IFD) coupled with energetic analysis that decomposes interaction energies at the atomic level. Using these tools we were able to elucidate the previously unknown substrate binding modes of a series of hydroxyurea analogs to human hemoglobin, catalase and the concomitant structural changes of the enzymes. Our results are consistent with kinetic and EPR measurements of hydroxyurea-hemoglobin reactions and a full mechanism is proposed that offers new insights into possibly improving substrate binding and/or reactivity.

Mechanism of Homogeneous Reduction of CO2 by Pyridine: Proton Relay in Aqueous Solvent and Aromatic Stabilization
Chern-Hooi Lim,1 Aaron M Holder1,2 and Charles B Musgrave1,2; 1. Chemical & Biological Engineering, University of Colorado at Boulder; 2. Chemistry and Biochemistry, University of Colorado at Boulder
We employ quantum chemical calculations to investigate the mechanism of homogeneous CO2 reduction by pyridine (Py) in the Py/p-GaP system. We find that CO2 reduction by Py commences with PyCOOH0 formation where: a) protonated Py (PyH+) is reduced to PyH0, b) PyH0 then reduces CO2 by one electron transfer (ET) via nucleophilic attack by its N lone pair on the C of CO2 and finally c) proton transfer (PT) from PyH0 to CO2 produces PyCOOH0. The predicted enthalpic barrier for this proton coupled ET (PCET) reaction is 45.7 kcal/mol for direct PT from PyH0 to CO2. However, when PT is mediated by one to three water molecules acting as a proton relay the barrier decreases to 29.5, 20.4 and 18.5 kcal/mol, respectively. The water proton relay reduces strain in the transition state (TS) and facilitates more complete ET. For PT mediated by a three water molecule proton relay, adding water molecules to explicitly solvate the core reaction system reduces the barrier to 13.6 - 16.5 kcal/mol, depending on the number and configuration of the solvating waters. This agrees with the experimentally determined barrier of 16.5 ± 2.4 kcal/mol. We calculate a pKa for PyH0 of 31 indicating that PT preceding ET is highly unfavorable. Moreover, we demonstrate that ET precedes PT in PyCOOH0 formation, confirming PyH0's pKa as irrelevant for predicting PT from PyH0 to CO2. Furthermore, we calculate adiabatic electron affinities in aqueous solvent for CO2, Py and Py•CO2 of 47.4, 37.9, 66.3 kcal/mol respectively, indicating that the anionic complex PyCOO− stabilizes the anionic radicals CO2 and Py to facilitate low barrier ET. As the reduction of CO2 proceeds through ET and then PT, the pyridine ring becomes aromatic and thus Py catalyzes CO2 reduction by stabilizing the PCET TS and the PyCOOH0 product through aromatic resonance stabilization.