Multi-scale Quantum Models for Biological Macromolecules


Figure 1: Quantum mechanically computed electrostatic potential surface of the polarized electron density of DNA in solution relative to the gas phase density. The figure was drawn at the Supercomputing Institute's Scientific Development and Visualization Laboratory using the GRASP program.

(Click the picture for a larger version.)
Recent years have witnessed exciting advances in three-dimensional macromolecular structure determination and other biophysical methods that have followed closely in the wake of the genome sequence project. These have added greatly to the wealth of knowledge about biological structure and function. At the same time, digital technology and high-performance computing capabilities have increased at an exponential rate and have positioned computational biochemical modeling at the forefront of one of the most exciting scientific challenges of the new millennium-modeling biopolymer interactions in a cellular environment.

A fundamental step toward this goal involves inclusive efforts of the computational biochemistry community to design new integrated models capable of working in concert to provide accurate atomic-level details about biochemical reactions. These methods need to span a broad range of spatial and temporal domains and model chemical processes at a hierarchy of theoretical levels. The development of new multi-scale models that meet these challenges will play a key role in the study of complex biochemical problems.

Figure 2: Quantum mechanically computed electrostatic potential surface map of the Mg2+ ion-bound hammerhead ribozyme structure in solution. The figure was drawn at the Supercomputing Institute's Scientific Development and Visualization Laboratory using the GRASP program.

(Click the picture for a larger version.)
Supercomputing Institute Principal Investigator, Professor Darrin York of the Chemistry Department at the University of Minnesota, along with his Quantum Modeling and Simulation group are developing and applying new multi-scale quantum methods to study large biological systems. These efforts include the design of linear-scaling electronic structure methods that allow large protein and DNA/RNA systems in solution to be studied with quantum mechanical models. Graduate Researcher Jana Khandogin is working to develop these methods and apply them to study quantum mechanical electrostatic potentials and polarization responses of biological macromolecules in solution (Figure 1). These methods complement ongoing studies in the laboratory led by Graduate Researcher Jill Johnson focused on metal-ion catalyzed phosphate hydrolysis reactions such as the self-cleavage mechanism of hammerhead ribozyme (Figure 2) using hybrid quantum mechanical/molecular mechanical methods.

Professor York, together with Supercomputing Institute Research Scholar Dr. Anguang Hu, have developed integrated multi-level quantum and empirical models that can be used in concert to construct realistic environments to study important chemical processes. As an example, Figure 3 shows a (UO2)2+(H2O)4 ion in an aqueous environment. This ion is important from an environmental perspective and is often found to interact with biomolecules. Heavy metals such as uranium require a special quantum treatment that includes relativistic effects, because the core electrons are moving at speeds comparable to the speed of light. The water molecules close to the ion (shown in blue) can be modeled using less computationally intensive density-functional methods, and the effect of bulk water (shown in red) can be modeled by a simple molecular mechanical force field. Consequently, modeling this system in a realistic solvent environment requires the integration of multiple levels of theoretical treatment, each designed to balance accuracy with system size and combined to provide an accurate description of a complex environment.

Figure 3: Multi-level quantum model of a solvated (UO2)2+ (H2O)4 ion. The (UO2)2+(H2O)4 complex was modeled with an all-electron scalar relativistic quantum method, the inner 10 Å shell of 145 waters were modeled with a gradient-corrected density-functional method, and the outer 10-12 Å shell of 1,154 waters were modeled by a molecular mechanics force field.

(Click the picture for a larger version.)
Future work in Professor York's Quantum Modeling and Simulation group will target phosphate hydrolysis reactions that are catalyzed by divalent metal ions-systems for which quantum "many-body" effects such as polarization and charge transfer play a significant role and for which conventional models are currently the most problematic.


 

This information is available in alternative formats upon request by individuals with disabilities. Please send email to alt-format@msi.umn.edu or call 612-624-0528.
 


HOME | QUESTIONS | FEEDBACK
Events | Links | People | Programs | Publications | Support | Welcome



URL: http://
This page last modified on  
Please direct questions or problems to help@msi.umn.edu  
Website related questions or problems should be directed to webmaster@msi.umn.edu
The University of Minnesota Supercomputing Institute does not collect personal information on visitors to our website. For the University of Minnesota policy, see www.privacy.umn.edu.
© 2002 by the Regents of the University of Minnesota