Our research program focuses upon the application and development of new computational tools that target organic and enzymatic catalyst design, alternative environmentally friendly solvent design, and drug discovery. Fundamental problems in organic and medicinal chemistry are probed, such as elucidation of enzymatic reactions, controlling enantioselectivity for chiral compounds, transition structure prediction, de novo design of high-affinity inhibitors, and origins of drug resistance. Obtaining quantitative success with large-scale quantum and molecular mechanical calculations involves the development of improved force fields, software, and methodology.
Background: The Acevedo group collaborates with experimentalists across the country with the goal of
developing inhibitors for the treatment of multiple diseases. Varied techniques are applied including: free energy perturbations,
docking, molecular dynamics, Monte Carlo, and ADME predictions.
(1) Collaboration with Prof. Jaehyuk Choi (Yale School of Medicine) - Development of B-Raf inhibitors exhibiting anti-melanoma properties.
(2) Collaboration with Prof. Zandrea Ambrose (University of Pittsburgh School of Medicine) and Prof. Patrick Flaherty (Duquesne University) - Development of anti-HIV compounds that target cyclophilins.
(3) Collaboration with Prof. Raj Amin (School of Pharmacy, Auburn University) - Development of partial agonists for PPAR-γ and PPAR-δ for anti-diabetic and anti-cancer properties.
(4) Collaboration with Prof. Angela Calderon (School of Pharmacy, Auburn University) - Development of anti-Malaria drugs targeting Plasmodium falciparum thioredoxin and glutathione reductase.
Solvent Effects and Catalysis
Background: The aim is to develop and use computational methods to gain physical
insight into how external factors (enzymes or solvent) enhance the rate or stereoselectivity
of chemical reactions. For many reactions the role of solvent has been assumed to be static,
hence its effect is basically thought to be a contribution of solvation energy to the total
free energy of the system. However, direct participation of solvent molecules may occur in
which a few critical solvent molecules bind to the transition structure and lower the
activation energy or an electric field created by the solvent changes the shape of the
potential energy surface. In extreme cases the reaction path itself can be perturbed,
especially when Lewis acids are involved. This reinforces the need for thorough studies
on the intermolecular interactions occurring between solvents, catalysts, and reactions.
Collaborations with experimentalists include:
(1) Collaboration with Prof. Holly Ellis (Auburn University) - Mechanism elucidation for flavin-dependent monooxygenase enzymes.
(2) Collaboration with Prof. Joan Hevel (Utah State University) - Mechanism elucidation for protein arginine methylation (PRMT1).
(3) Collaboration with Prof. Doug Goodwin (Auburn University) - Enlisting peroxidatic electron donors to expand the catalatic activity of KatG.
Background: Ionic liquids are a novel class of solvents, defined as a material containing only ionic
species, with a melting point at or below room temperature. In sharp contrast to molten
salts or melts, ionic liquids can be fluid at temperatures as low as 204 K, are colorless,
have low viscosities, high conductivity, negligible vapor pressure, excellent thermal and
chemical stabilities, are recyclable, non-explosive, easy to prepare, active at room
temperature, and tolerate impurities such as water. An exciting aspect of ionic liquids resides in their ability to provide increased
rates and selectivity for a series of industrially and academically important reactions such
as the Heck, Friedel-Crafts, isomerizations, hydrogenation, organometallic, Michael, Mannich,
Wittig, 1,3-Dipolar additions, aldol and benzoin condensations. The observed effects of ionic liquids range
from weak to powerful, but an understanding of the molecular factors are largely unknown.
Objective: The objective is to understand the microscopic details on how ionic liquids operate and to exploit this understanding to predict new ionic liquids that give optimal rate and stereoselectivity enhancements. A comprehensive understanding on how ionic liquids impact chemical reactivity will be used to influence a wide range of (1) difficult organic reactions, which benefit from toxic solvents coupled with high pressures and temperatures, and (2) enzymatic reactions, which require complex physiological conditions. The long-term intent of our research program is to create controllable, efficient, safe, and environmentally clean technologies that impact society and chemistry from the laboratory bench top to large-scale industrial manufacturing.
Collaborations with experimentalists include:
(1) Collaboration with Prof. Renato Contreras (Universidad de Chile) - Chemical reactions in ionic liquids.
Background: A GPU (graphics processing unit) is a graphic chip that processes the visual
display in a computer. Over the years, these graphic chips have dramatically increased in computing
power due to high demand from the consumer videogame market. As a result, GPUs have become very
attractive for general purpose scientific and engineering computing. GPUs cost a fraction of a
regular computer processing unit (CPU) and can provide dramatic enhancements, e.g., 100x the
processing power per dollar spent. Consequently, 3 of the top 5 supercomputers in the world
are NVIDIA GPU-based computers. The major disadvantage is that software must be rewritten to
utilize GPUs. NVIDIA has introduced a massively parallel architecture called "CUDA" which is
fully programmable for scientific applications, but requires expertise.
Objective: The objective is to routinely simulate large systems containing millions of atoms and thousands of residues. We propose the creation of a new program, MCGPU (Monte Carlo on Graphics Processing Units), to take advantage of the dramatic power of the GPU. Development of MCGPU has begun and individual subroutines have been completed. For example, functions to run a Monte Carlo simulation on a box of solvent molecules using NVT and NPT ensembles and periodic boundary conditions have been coded. The CUDA enabled code can perform a speed up of easily an order of magnitude relative to the single-processor CPU code. When a stable version of MCGPU is completed, is will be distributed as a freely-available, open source program for academic groups.
• NSF, National Science Foundation (PI: CHE-1149604; Co-PI: DEB-1244320)
• Alabama Supercomputer Center
• Auburn University