Physical Biosciences Division at Berkeley Lab

Our department develops quantum and statistical mechanical approaches to elucidate complex phenomena pertinent to systems of pragmatic importance. Within this central theme, we focus on four broad areas: cell-cell recognition in the immune system, polymer science and engineering, sensor technology for pathogen detection, and heterogeneous catalysis.

Computing resources at Berkeley Lab's NERSC

Understanding how cells respond to external stimuli is an important goal in biology. The response to external stimuli is the result of interactions between many cellular components, and the spatio-temporal evolution of these components can now be vividly observed in imaging experiments. However, it is difficult to intuit the mechanistic underpinnings of phenomena from these observations alone. This is because the observed complexity emerges from collective dynamics of the many interacting cellular components. The goal of our program is to develop multiscale theoretical and computational approaches that serve as full partners of genetic and biochemical experiments in the discovery process in cell biology. To take steps toward this goal, we are currently studying various aspects of T lymphocyte activation and the behavior of some eukaryotic cells using synergistic computational studies and genetic, biochemical, and imaging experiments. Understanding these systems is of direct relevance to biomedicine and energy related technological applications.

Protein folding and structure prediction
The key to understanding the inner workings of cells is to learn the three-dimensional atomic structures of the full repertoire of macromolecules that form their architecture and carry out their metabolism. These three-dimensional (3D) structures are encoded in the blueprint of the DNA genome. Within cells, the DNA blueprint is transcribed into RNA and translated into protein through exquisitely complex machinery- itself composed of proteins and RNA. The experimental process of deciphering the atomic structures of the majority of cellular proteins is complemented by new algorithm developments and advances in computer hardware that will learn to decipher the DNA message by computer. Within the Computational Structural Biology group in the Physical Biosciences Division's there are individual research programs in the areas of protein structure prediction and folding, RNA structure prediction and classification and RNA gene discovery, and computational tools for macromolecular structure determination and analysis by x-ray and cryo-EM.

At the crossroads of chemical engineering, chemistry and physics

Arup Chakraborty named Department Head

more news from PBD...

Collaborators

Dr. Paul D. Adams, Deputy, Staff Scientist, Physical Biosciences Division, LBNL
The Computational Crystallography Initiative (CCI) is part of the Physical Biosciences Division at Lawrence Berkeley National Laboratory. The focus of the initiative is the development of computational tools for macromolecular structure determination and analysis. We are developing a new software system, called PHENIX, as a flexible platform for these tools. Collaborations between members of the CSB group and the Physical Biosciences Division will introduce new methods into PHENIX, such as structure determination by electron diffraction, and macromolecular simulation using molecular dynamics methods.

Dr. Robert M. Glaeser, Faculty Staff Scientist, Physical Biosciences Division, LBNL

Computational Interests: Automated and accelerated processing of single-molecule images by electron cryo-microscopy. High throughput structure determination at high resolution.

Methods: Computational recognition (identification) of identical particles that are presented in alternative views. Alignment and orientation-determination followed by merging data from 105 to 106 particles to improve signal to noise and resolution.

Stephen Holbrook Dr. Stephen R. Holbrook, Staff Scientist, Physical Biosciences Division, LBNL
Computational Interests: Protein structure prediction, RNA structure prediction, molecular modeling, structure and sequence databases, RNA gene discovery, protein function prediction, development of methods for crystallographic structure determination.

Methods: X-ray crystallography, machine-learning methods (neural networks, support vector machines, hidden Markov models), protein and RNA preparation, purification and synthesis. Structural Classification of RNA (SCOR) database (co-investigator - S. Brenner), RNA Gene Prediction

Dr. Teresa Head-Gordon, Faculty Staff Scientist, Physical Biosciences Division, LBNL
Co-editor of Community White Paper on Computational Biology
Dr. Head-Gordon's research group combines experimental, theoretical and computational approaches used toward the problems of protein folding and structure prediction and modeling aqueous hydration. The research is highly multidisciplinary and collaborative among the areas of structural biology, physical chemistry, physics, computer science and mathematical optimization.

Computational Interests: Hydration forces in folding of proteins, minimalist protein folding models for annotating whole genomes, global optimization approaches to protein structure prediction, models of the condensed phase using ab initio molecular orbital theory, modeling chemical bonding effects for electron crystallography, new simulation methods.

Methods: Molecular dynamics, Monte Carlo, electronic structure theories, global optimization, diffraction and small-angle solution scattering experiments.

Home \| Science \| Facilities \| About Us \| Employee Resources