Computational Biology

The ORNL Computational Biology Program consists of members of Life Sciences Division, Computer Science and Mathematics Division, and Environmental Sciences Division. It is oriented toward various aspects of genome and proteome analysis, molecular systems biology. Elements of the effort include:

Genome Analysis and Annotation

Genome sequence analysis systems including large scale pipeline for annotating microbial and eukaryotic genomes, including environmental microbes sequenced by DOE, human, mouse, poplar tree, Ciona intenstinalis and others. The tools in these systems consist of genefinders, sequence comparison systems, protein motif and domain detection systems, regulatory element detectors, and others. Large scale annotation is provided for a variety of microbial and eukaryotic genomes by the Genome Analysis and Systems Modeling Group (http://compbio.ornl.gov/ ). Systems for presentation and query of genome analysis results can be found at the DOE Joint Genome Institute (http://www.jgi.doe.gov/). These provide views of genomic features and permit user retrieval of linked information from genes to proteins to metabolic maps and gene function classification data. http://compbio.ornl.gov/public/tools contains analysis tools for computational biology.

Databases, Data Integration and Collaboratory Systems

In the area of bioinformatics, significant efforts are focused on the development of laboratory information management systems, statistical and clustering approaches and tools for the analysis of expression data, and collaborative environments for bioinformatics and biology. Data Integration projects are being developed in areas of microbial functional genomics (Shewanlla Federeration - http://) Mammalian genetics and genomics (http:// , and The department of homeland security - http://). The Tennessee Genome Mouse Consortium (TMGC), which brings together expertise and resources across the state of Tennessee for the study of complex biological systems using mouse model systems ( http://www.tnmouse.org).

Molecular Modeling and Simulation

Protein threading and folding efforts include tools such as the PROSPECT threader, which is an online server, systems for molecular dynamics on high performance computers, and codes for using various types of experimentally obtained data such as NMR data or scattering data, to assist protein fold recognition and modeling. http:// Also, projects using molecular dynamics to study protein vibrations and enzymatic function are being facilitated using high performance computing infrastructure (Pratul Agarwal - agarwalpk@ornl.gov).

Recent efforts have focused on developing methods to accurately predict docked molecular complexes using high performance computational searches and with constraints from experimental data from such techniques as residual dipolar coupling, and chemical cross-linking. This development is keyed to the DOE Genomics:GTL program which lists understanding molecular machines as one of its major goals.

Mass Spectrometry Analysis

Complementing this GTL work are methods for analyzing mass spectrometry data, particularly tandem MS data of proteins. The new methods developed include recognition of ion types in MS/MS spectrum, computational parent ion charge determination, advanced statistical scoring functions for peptides and proteins, and theoretical fragmentation 'models. The MASPIC scoring system for MS.MS analysis can be downloaded at http:// . Tema, MASPIC, PPM

Metabolic and Regulatory Modeling

Methods for cell network modeling have been developed and are an active area of research including flux-based metabolic models and stochastic regulatory network simulations.

High Performance Biocomputing

ORNL has a Center for Computational Sciences and National Leadership Computing Facility which supports infrastructure for scientific application in biology and other fields (http://nccs.gov/). The Computational Biology Institute, within the CCS and Life Sciences Division, is a matrix for computational biology activity related to large-scale computing applications ( http://www.ccs.ornl.gov/cbi/). A new project from the Office of Advanced Scientific Computing which seeks to develop highly efficient codes for mass spectrometry analysis, molecular docking, and network simulation on large shared memory computers. (http;//)

Computational Biology Graduate Training

The Graduate School in Genome Science and Technology (http://gst.ornl.gov), offered by The University of Tennessee and the Oak Ridge National Laboratory (ORNL), is a unique, multidisciplinary program for full time graduate study leading to an M.S. or Ph.D. degree in the emerging field of genome science and technology. One of its foci is computational biology/ bioinformatics.

(Contact: Ed C. Uberbacher, ube@ornl.gov, 865-574-6134 )