DATE: April 27-28, 2000

LOCATION: American Geophysical Union, Washington, D.C. The meeting was announced in the Federal Register for November 30-December 1, 1999 (on April 7, 2000, Volume 65, Number 68, Page 18320).

PARTICIPANTS: A list of attendees showing all BERAC members who were present, guests, and participating Department of Energy officials and staff is attached.

Dr. Jim Decker – Acting Director Office of Science – FY 2001 Funding Outlook

• FY 2001 budget looks quite good with overall growth for the Office of Science (SC).

• On the surface it appears that the BER FY 2001 request only has a 3% increase compared to FY 2000; however, this is misleading because the FY00 base includes over $30+M in FY00 earmarks, so it is actually much better than it looks.

• Nanoscale science, engineering and technology – building structures one atom at a time. This exciting BES initiative is part of a broad administration initiative that is requesting ~$500M across several agencies. NSF is the lead agency. The FY01 request would double the FY00 investment. Nanoscale science is viewed as having enormous technological payoffs.

• Scientific computing is an important area across all SC programs. SC had a significant scientific computing request for FY00 (the Scientific Simulation Initiative – SSI) that was not funded. This failure resulted, in part, from a funding shortfall and from significant criticism of the proposed effort. DOE was asked to develop a non-defense program supercomputing plan. This has been submitted to Congress. The FY01 request is for a very different program. Last year the big part was for hardware. In FY01 the principal focus is on global climate change and combustion with $20 million for SC programs, including $8 million for BER. The hardware request is down significantly. One proposed goal is to upgrade NERSC to ~5 Tflop. Overall, the goal on the hardware end is to make tools that will be useable in the future, since it is currently difficult and a real challenge for many people to use our super computer resources.

• Biomedical engineering – use of the laboratory’s multidisciplinary capabilities. This is not a big request for us at $6.6 million. This DOE program is a difficult sell on the Hill. NIH has so much money in this area, we have been and will be asked why DOE should do anything in the biomedical area? One of the strongest arguments to be made is that U.S. biomedical research has benefited historically from multiple funding sources.

• Several recent accomplishments – the Fermi Lab main injector was completed in FY99. The SLAC B factory and RHIC are now in full operation. CEBAF also had a good year.

• Spallation Neutron Source – We have a new management team in place at the SNS now. We just completed a Lehman review at the SNS and did well. We are now confident that the project can be done on cost and on schedule. This will be the premier neutron scattering facility in the world when it is completed.

• Scientific facilities utilization – There is a renewed commitment in the FY01 request to the FY96 presidential facilities utilization initiative. The goal is to increase the number of users at our facilities and to make increased investments to improve their quality.

• E&W subcommittee $ currently $750M below request and 200-300 below FY00, people assume there will be a fix, but will likely be a tough year especially since anxious to get out of town.

Dr. Jae Edmonds – Battelle Pacific Northwest National Lab – Global Change Research

• It’s a fossil fuel world – only about 12% of our current energy technologies are noncarbon emitting.

• Essentially every projection shows oil and gas resources ending early in this century. But there is still lots of coal out there. Without proactive efforts we could easily double the amount of CO₂ in the atmosphere.

• The recent Intergovernmental Panel on Climate Change (IPCC) report makes projections of various levels of atmospheric CO₂ stablization levels under different CO₂ emission scenarios, The bottom line is that CO₂ emissions need to be controlled. However, we don’t necessarily have to phase out fossil fuels. There is a whole suite of technologies available that can help.

• What if we implemented the Kyoto protocol? If everyone did what they are supposed to it actually wouldn’t result in that much of a drop in CO₂ emissions. We would only get a few ppm reduction.

• Stablization of CO₂ levels actually requires return to 1990 emission levels. With the world population doubling, this would require a halving of carbon use per person. Some of the scenarios that have been put forward involve some substantial challenges, e.g., a 50 mpg car by mid-century, 75% electricity from carbon free sources compared with approximately 25% today.

• Technology is key - energy technologies, terrestrial carbon repositories (forests, soil, etc.), carbon capture technolgy, carbon sequestration, etc.

• How much CO₂ emission reduction is needed? 10-50% depending on the desired ceiling, i.e., 350-750ppm.

• Lots of technology options will be considered that wouldn’t otherwise be considered because they are expensive.

• There are a variety of very different technology needs to be considered depending on assumptions that are made about future energy source options.

• Most research on various energy options is being funded in the U.S. and Japan. There is a very different mix of research that is being funded even in these two countries, e.g., conservation (Japan), fossil (U.S.), fission (Japan), basic energy technology (U.S.).

Dr. Ray Gesteland – Howard Hughes Professor, University of Utah - Chair, BERAC Genome Subcommittee

Introduction to minisymposium addressing charge to BERAC on the future of BER biological research given progress in genomics

• The subcommittee has developed a draft research program plan that focuses on how cells run their daily lives and how they respond to their environment.

• This program would require a coming together of the national labs and would require that DOE attract the best from academia and the commercial sector.

• This multidisciplinary research effort would require a much more serious coming together of physical, chemical, biological, and computational scientists than we have even seen in the human genome program.

• Although the discussions at the subcommittees three meetings to date covered a lot of ground the focus kept returning to being able to understand biology well enough to be able to predict the behavior and response of biological systems – from cells to organisms. This type of predictive capability would also serve as a "hypothesis generator" to validate our understanding of biology.

• As biology moves forward in what some are calling the post genomic era, i.e, after the human genome has been sequenced, it is important to realize that DNA sequencing needs will continue to grow rather than end.

• Key research areas:
  1. focus on genes, the regions of DNA that regulate gene expression both quantitatively and temporally, and the proteins that interact with these regulatory regions (such an effort will require considerable DNA sequencing);
  2. global gene expression issues, especially at the protein product level rather than the messenger RNA level that is most frequently used today;
  3. real time monitoring of gene expression in single cells and tissues.

• Each of these research areas would require a solid grounding in genetics and biology. The mouse will be one of the most useful models.

• What are the scientific opportunities? What are the technical challenges? What is the role of DOE? How does this broad research agenda map to the DOE mission? What are the management requirements? What are the training needs across scientific disciplines?

• This new view of biology represents a move back from reductionism to systems level biology. It will require new types of scientists. How do we engage labs and academia to get these specialists? What are the ethical, legal and social issues associated with the new knowledge and biological capabilities that will arise from such research.

• The following series of five mini seminars are intended to provide examples of the opportunities, needs, and benefits that lie ahead of us in biology as the number of organisms, including human, whose entire genomes have been sequenced. These mini seminars will be followed by an extended open discussion that will be used to guide the next phase of the genome subcommittee’s deliberations to develop a draft program plan for broad scientific input and for presentation to BERAC.

Dr. Barbara Wold - Professor, California Institute of Technology

"From Sequences to Biological Regulation"

• So far, most biologists have tended to emphasize the "building block regions" of the genome – the genes. This is especially true in our reporting of progress in sequencing the genome, i.e., how many new genes have been discovered.

• The DNA sequence is really more of a recipe for life than it is a blueprint since it contains the parts list plus the rules for how those parts work.

• Living systems likely employ very different design rules than the engineered systems that we now design. Genes operate and have their varied effects through differential gene expression.

• Biological diversity – the scope of the problem. By histology there are about 400 distinct cell types in an adult human. If we count the different stages of development and the diversity seen in neural tissue there may be a few thousand distinct cell types. Even with 100,000+ human genes, there simply aren’t enough to account for all of the biological diversity within one individual. Diversity comes from changes in the physiology and environment of cells, tissues, and organs. It includes combinations of and changes in the amounts of the genes being expressed at any one time and over time. To understand how all of this happens and works will, in the end, require models that can incorporate both biology and complexity.

• For years, biologists have used a one-at-a-time approach to studying how genes or small networks of genes work. Not only do we not want to do biology like this any more, we simply can’t afford to.

• Big 5 Year Goals:
  • Identify all functional cis-regulatory elements for all genes – very doable for at least 90% of all genes in an organism. An essential and very powerful tool is comparative genomics – the comparison of the genomes of human, mouse and another organism. Selecting the right genomes is very important. They need to be far enough apart so that genetically important DNA sequence information is conserved while other sequences have diverged enough so that they are clearly different. This approach provides tremedous biological sifting power.
  • Define all gene ensembles that are co-regulated by related cis-elements – also very doable.

• What is the problem in being able to do this? Can’t just read this information in the DNA. Functional assays are good but very slow and don’t provide information at the genomic level. Example - Eddy Rubin at LBNL has studied the mouse/human interleukin gene cluster. He didn’t know much about this region to start with. He compared regions of the mouse and human DNA sequence where homology was high - some of these were coding regions and some were not. Blocks of human DNA were tested in genetically engineered mice (where one can test for the gain or loss of a specific function) to determine the effect on neighboring genes of a conserved, noncoding human DNA. Some were affected and some weren’t, providing a link between a regulatory region and the genes that it regulates.

• Just having the complete DNA sequence for human (or any other organism) is not the answer. However, comparative sequencing of different organisms is a great hypothesis generator.

• How can comparative genomics take a major step forward? This approach is very powerful if done right by selecting the right organisms for comparison. For example, mouse and human are usefully positioned about 100 million years apart in evolution. We would like to have a third genome with about 100 million years separating it from both human and mouse – a so-called Golden Triangle of Genetic Drift. A good choice might be an organism that would never be used as an experimental organism – perhaps the rabbit or dog. The chicken also has pretty small/compact genome.

• How do you pick organisms to fill-in a Golden Triangle? Pick several organisms and representative questions to be addresses and do some quick sequencing comparisions and/of tests. In the end, we may want to have more than one of these Golden Triangles to focus on specific problems/questions. Genes work in circuits. There are only a few circuits for which we know initiating regulators, co-regulators, inhibitors, and target genes, e.g., muscle making circuit and myoD as a master regulator. Turns out there are lots of cross connections with other circuits. What starts the circuit running – there is input from many different sources. Comparative genomics can help define who talks to whom and what regulates the regulators? Studying whole genomes lets you look at whole circuits not just pieces.

• Central 5-10 year goals
  • identify regulatory proteins that interact at each cis-regulatory sequence (doable)
  • define the function of (silencers, enhancers, etc.) (doable)
  • do all of this at a genomic scale

• Central 10+ year goal
  Build modeling vocabulary -
  • to predict gene circuits and their higher order integration
  • predict biological behavior when a system is perturbed – doable now at some levels, but our greatest challenge

• Models are required to handle complexity. They are a key component of our movement from individual genes or gene circuits to genomic scale biology.

Claire Fraser - President; Director, The Institute for Genomic Research

"From Sequences to Gene Function"

• Major challenges of going from DNA sequence to gene function:
  • 40-50% of putative genes identified in genomic DNA sequencing projects to date are of unknown function even though we have completed 30 genomes. In addition, 25% of unknown coding sequences are still unique, i.e., aren’t being found in other genomes being sequenced.
  • Expectation from sequencing projects over the next 2-3 years – 350,000 microbial genes will be identified, 150,000 of these will be of unknown function, and 20-50K of these genes will be novel.
  • A variable (and in some cases a high) fraction of archeal genes are being found in other microbes (up to 25% in some cases!). For example, thermatoga has 24% archeal genes while E coli only has 2.4%. These examples illustrate the value of having complete genome sequence information. Understanding and studying evolution requires genomic scale analysis versus gene/gene family studies.

• How do we get at functional genomics? Many options – expression analysis, genomie engineering, minimal gene set analysis, etc.

• Expression arrays will continue to play a big role but they have a number of limitations. They are not widely distributed yet and are still relatively expensive. Should we think about developing centralized facilities for these kinds of analyses? Sensitivity is also an issue below 2-3x differences in mRNA expression. Low abundance messages present problems. At best there is only a 50% correlation between RNA levels in cells and the levels of the corresponding proteins. There is a clear and urgent need for large-scale proteomics techniques.

• M genitalium as an example of a model organism to help define a minimal gene set needed for independent life and, possibly, to define biological responses to environmental stimuli. Studying an organism with a minimal number of genes may be less ideal than initially thought since there is no "biological buffer." We need model organisms with enough genetic redundancy so that they can actually respond to environmental stimuli. When M genitalium was exposed to a variety of environmental conditions either all of its genes were expressed or the cell died.

• There were also some unexpected effects of knocking out genes that we thought we knew what they did. This approach to understanding and defining a set of genes needed for life will be tougher than expected.

• New Challenges & Opportunities
  • DNA sequencing needs – especially across unsampled phyla
  • Expanded functional genomics studies
  • Better methods for microbial manipulation
  • Annotation tools/methods
  • Tools for linking data from functional studies with sequence databases

Keith Hodgson - Professor, Stanford University; Director, Stanford Synchrotron Radiation Laboratory; Chair, BERAC

"From Sequences to Proteins"

• Focus here on synchrotron enabled structural biology in the post genomic era – DOE and its unique role in the emerging biosciences.

• Synchrotron radiation has revolutionized structural biology in the world of crystallography. The "world record" to go from a protein to a structure has gone from man years to minutes.

• Examples of remarkable molecular machines whose structures and inner workings have been uncovered because of current capabilities in structural biology:
  • RNA polymerase II (a 450kDa cellular machine on the cover of an upcoming issue of Science);
  • the ribosome (a 2.5MDa machine with >55 proteins and 4500 nucleotides that was on the cover of Science last year);
  • both of these newly uncovered structures give unprecedented access to the function of these molecular machines

• New challenges and opportunities in structural biology:
  • Membrane bound proteins represent another grand challenge in structural biology
  • Opportunities in kinetic synchrotron crystallography – "structural snapshots and movies of protein behavior with time"
  • Developing a collaboratory environment for structural biology research. Scientists working at distributed facilities without leaving their home laboratories.

• Broad goals of structural and functional genomics – comprehensive database of protein motifs and folds and functional relationships based on biochemical analyses.

• Lessons from genome sequencing
  • high throughput methods are key
  • importance of multi-institution collaboration
  • key role of computation for data analysis/mining

• Rapidly growing/developing programs in the public and private sectors

• Example of multi-institute collaboration – labs, universities, industry – that cut across all areas of need in structural genomics:
  • high-throughput protein production example (robot at Novartis)
  • high throughput crystallization, nanovolumes, 50-100 fold reductions in protein
  • high throughput data collection and analysis – goal of a few structures per day (big number compared to where we have been fairly recently though small compared to DNA sequencing)

• Key challenge - Structures of super assemblies (molecular machines). Details of function at the chemical level will contribute to understanding of cell behavior/response.

• Key strategy - Aim for activities that are out of reach of individual investigators or small teams. Interagency coordination, novel management needs, foster new technology innovation and transition to use in production-scale experimental approaches

• Key challenges - Structural genomics is always presented as a pipeline but there are some constrictions. Protein production is a nontrivial challenge. Sorting out and understanding macromolecular interactions has always been difficult. Protein crystallization is probably the easy part.

• Key challenge - Linkage (or the current lack of linkage) between the different areas of experimentation, genomics, structure determination, etc. Currently proposed centers (to NIH solicitation) have each chosen their area of emphasis – whole organism, biochemical pathway, etc. How, in the long run, do we define the priorities for access to these types of facilities? Peer review at some level.

Lisa Stubbs - Senior Scientist, Lawrence Livermore National Laboratory

"From Sequences to Organisms"

• A key challenge is to translate DNA sequence to gene function in vivo – we can’t do this today.

• Commonly we use mutations to study gene function – induced, knockout, disease families (only ~7,000 of ~140,000 human genes have actually been characterized this way). This is only the tip of the iceberg. Global information on gene fucntion isn’t going to come from human studies.

• More and more mutations are being induced in mice. The results can be extrapolated directly to humans. Genome-wide mutagenesis approaches are being used. DOE has a long and strong history in mouse genetics and mouse mutagenesis.

• A variety of mutagenesis schemes can be applied efficiently in mice –
  • Gene knockouts delete a specific gene, regulatory sequence, or genomic region
  • Gene replacement - replace a normal mouse gene with a modified copy or a human gene
  • Transgenics have human gene regions introduced to study their function and/or regulation in a model system (e.g., system being used by Rubin at LBNL).
  • Combined methods - targeted mutations such as deletions plus chemical mutagenesis produces new mutations that saturate genes in a specific region (e.g., system developed by Woychik at ORNL).

• Mice are expensive to use in large numbers but they are irreplaceable for many applications simply because they are mammals. Many processes most relevant to human health can only be modeled in mammals. Other organisms can be used for some specific applications. Mice also have the advantage of the many inbred strains that have been developed. Using inbred mice is like having large sets of identical twins. Mouse biology also has a rich history of research in genetics, pathology, embryology, etc.

• The effects of mutating specific genes are often surprising demonstrating how little we actually know about the complexity of biological systems.

• Key challenge - Assembling a molecular parts lists for a cell and modeling entire cells, even simple ones, will require the collection of a massive amount of basic data on gene regulation, molecular interactions, biological pathways (metabolic, developmental, networks), development of structures, cell-cell interactions (communities to tissues), etc.

• Mutations only define the endpoints of affected pathways not the pathways themselves. They also provide crude tools to test hypotheses about biological processes.

• Key opportunity - We should use models that suit our needs - from microbes to simple multicellular organisms (for basic developmental information) to vertebrates (for higher order developmental events, e.g., zebra fish) to the mouse.

• Key challenge – Defining a biological parts list and putting the pieces back together:
  • biological endpoints shared by and distinguishing different types of organisms, e.g., sampling across microbes, fungi, vertebrates, mammals
  • defining overlaps and differences in the parts lists of strategic species
  • correlating parts lists to sequence conservation and differences
  • develop modeling techniques that can reassemble these complex processes from sequence data

• Key challenges/opportunities –
  • new tools for manipulating genes in microbes and other systems
  • mice – targeted mutagenesis is costly and tedious; are there better approaches?
  • whole genome mutagenesis is efficient but linking genes to mutations will still be daunting and expensive given the need for a baseline of 140,000 mutants to understand the human genome
  • alternative models should be explored to model some types of conserved functions

• DOE role -
  • Microbes – development of methods to examine gene function; sampling the evolutionary tree to find the parts list
  • Mice and other model organisms – Mutants will be the essential tools to test and develop models of gene function. DOE has a tradition of investment and unique expertise in mouse mutagenesis and genetics. Methods of linking mutations to genes are still primitive.
  • Informatics and modeling – The tools we will need soon do not exist to keep track of information and to put it back together.

Adam Arkin - Staff Scientist, E.O. Lawrence Berkeley National Laboratory

"From Sequences to Biological Models and Predictions"

• Cells and other engineered systems have roughly two tasks: perform a particular function and do so robustly in the face of uncertainty. Both are associated with issues of control versus function.

• Wings will actually fly by themselves in wind tunnels but in reality, to have flight, there is a need to deal with far more variables than even exist in a wind tunnel.

• Microbes have on the order of 10¹⁰ molecules representing ~4000 different types of molecules. We don’t currently have models or understanding for even the simplest cells compared to the level of understanding that we think we have for the behavior and/or function of complex machines.

• The integrated circuit revolution began in ~1970. We needed theory and computational methods to deal with the added complexity. We know all the parts. We know the "device physics." We know the interactions. But we still don’t fully understand how an integrated circuit works. In contrast to comparable mechanisms in cells we don’t even know the parts, the physics, the interactions or how it works.

• Modeling forces you to think about all your data to ensure that it is maximally consistent. It enables testing of biology. Even the simplest biological systems and processes are very complex.

• Key challenge - There is a great diversity of biological data available but most data is "in the hands of the biologists," i.e., not necessarily as consistent or as useful as modellers would like it to be.

• Key challenge - What’s the dream? Computation is not nearly as costly as experimentation. We would like to find the parts of biological systems, to determine the physical properties of those parts, and to understand how those parts are integrated and work together as parts of cellular networks.

• Key challenge - Quantitative analysis important. How do you decompose the individual elements of a network into pieces that can be reassembled without having an overall knowledge of the network itself? The behaviors of a biological circuit can vary depending on biological and "environmental" conditions/variables.

• A Simulation Program For Integrated Circuit Evaluation (SPICE) has been developed as an engineering tool. We are developing comparable tool(s) for biology - BIO/SPICE – that includes elements of databasing, simulation, and analysis.

• Just because you have a lot of parameters in your model doesn’t mean you can explain a lot of things!

• Key challenge - How do you choose the problem(s) to solve? There isn’t much data out there that is really very useable and most of the data that does exist is far too complicated.

• Key opportunity and challenge - Choosing a (microbial?) cell for study. A representative of a wide range of microbes? Interest to many users? Interest in nearby microbial relatives? Wide variety of behaviors? Evolutionarily distant cousins to compare to? Experimentally workable? (growth, genetics, modification capabilities?)

General Discussion – BERAC and public participants

• Ray Gesteland introductory comments:
  • The subcommittee has already met on several occasions and has put together a very rough draft proposal.
  • Following the input from this meeting, this draft proposal will be put into a form that will be sent out to a broader cross section of the scientific community for comment.
  • The plan is for this proposal/recommendation to be in a form that can be approved by BERAC before the next meeting in November.

• We need to ensure that there is an active linkage between modeling and experimentation. We have to be modeling as we collect the data.

• New scientific/research interfaces need to be identified and nurtured.

• There is a nucleus of people within the labs who have multidisciplinary training. What are these people doing now and how do we get them to do other things? We need to figure out how to scale what is already happening in individual labs to a broader group. This will require considerable effort, commitment, time, and communication.

• DOE has an ability to put together infrastructure where people will want to come and do research. Modelers like being at the mouth of a data stream and would fit very well as part of a broad infrastructure to address the kinds of questions being asked here. Biologists like having access to resources they can’t get other places.

• What does biology and the biologist of the future look like? We need a combination of new ways of doing biology combined with the contribution that this new biology makes to DOE needs/problems. The biology being done in 5 years will not be recognizable from today’s perspective – this is a hard thing to see today.

• We need focused teams of scientists with common goals versus each scientist doing their own thing. A Joint Genome Institute type model is needed – centralized facility versus distributed facility? Need for co-localization of some things but not others.

• More than just a lot of resources is needed make all of this happen. What is needed is almost a SWAT team approach to biology. We may need to start from scratch and bring groups of scientists together to achieve something of this magnitude. It is impressive to look back on the JGI experience to see why it worked so well. Almost a controlled panic mode of operation. High value of external/independent oversight. Scientific and management discipline and flexibility.

• To motivate people to get involved we need a win-win situation. Something analogous to the ramp or die motto of the JGI used to describe the scale-up of their sequencing operation. The challenges here are far more complex and the path is far less clear.

• We need a goal/program that is so compelling that people won’t be able to resist.

• Draft Goals -
  • Understand the internal function of cells through their genes and gene products
  • Understand the interactions and communication within communities of cells
  • Predict biological responses and human susceptibility to external stimuli
  • Adapt the sophistication of biological systems to man-made systems and machines

• What is the next target we are aiming at? Something you can go to the Hill with?

• These are all the right things to do. What is the link/justification within the DOE mission? These are things that all biologists will be doing anyway. What is DOE’s niche?

• We can’t assume that we know which proteins are involved in susceptibility to energy related materials for example. Need to contribute to general knowledge while keeping underlying DOE goals in mind. Clearly things like carbon sequestration are much more straightforward in terms of the organismal targets selected for study (but not necessarily at the molecular level) though the tools and resources needed are the same in many cases

• We don’t even know how to construct the databases that will be needed.

• Nucleus around which to build a program - several possibilities: computation, modeling and databases as central organizing principal. You can’t pick targets for study too specifically. For example chromosome 19 was originally chosen as a DOE target for sequencing because of the repair genes located on chromosome 19; however, we obviously we didn’t just sequence the repair genes.

• DOE does have a mission to provide/develop facilities that others can’t provide for themselves. The kinds of resources being talked about here clearly fit that model.

• Is this anything different or just more of the same that everyone else will be doing? DOE can/will certainly take a more global approach to the biological questions being raised in contrast to the "cottage industry"/single investigator approach that typifies NIH and NSF programs. DOE would likely also make a choice of genomes, e.g., a microbe with potential value for energy or environmental uses, for detailed characterization that would be unique/different from the ones others might choose even if the overarching questions, e.g., discovering all regulatory elements/molecules, were similar.

• NIH is an applied mission agency that works from first principles. The same kind of words can be found in the enabling legislation for DOE, i.e., solving problems from first scientific principles.

• DOE is already embarking on big investments in computing and the development of a neutron source. This new effort needs to impacts these ongoing DOE investments. For example, biologists need to have an impact on what happens in computing. Biology can be left out just as easily as it can be a driver.

• Would some aspect of science not get done if BER did nothing here? Would anyone say that biology is in trouble because BER isn’t doing their thing? Haven’t heard these kinds of compelling arguments yet.

• There is a lot of biology (including modeling) that very difficult to get funded through NIH. The systems approach to biology with a focus on modeling is a powerful driver. Novel facilities are needed to generate the new data - data sets that would never get produced by the cottage industry approach to biology.

• There are ongoing computational efforts at NIH and NSF. (See NIH Biomedical Information Science and Technology Initiative report – BISTI - http://www.nih.gov/about/director/060399.htm) for information on NIH efforts. This effort is based on a recognition of the capabilities and facilities in the NSF and DOE programs (~$200M at NIH, ~$500M at NSF).

• Training opportunities/possibilities? Some universities have started interdisciplinary programs that involve informational/computational sciences. Training programs for established investigators versus the more typical focus on students? There is a great deal of interest in this since it would serve as an instructional core to support the next generation of students. It should be noted that the Advanced Scientific Computing Initiative (ASCI) included training as an important aspect. There is currently no shortage of biomedical grad students and postdocs, they are just being trained to do what their mentors already do. Could training programs at the national labs to do some of this? What about distance learning opportunities?

• We need analytic technologies to develop and study data that simply don’t exist today? This needs to be included as aspect of report/recommendation.

• This still sounds a lot like all of biology for the next decade - too broad. Too big to capture in a single initiative? The program could have microbes as a defining feature plus some of the complementary aspects of the human/mouse studies.

• Traditionally BER has had two biologies – environmental biology and basic biology related to susceptibility/radiation/etc. These are now coming together in terms of needs and opportunities.

• There is considerable work going on involving the hybridization of biological systems with machines. For example people are trying to couple neurons with semiconductors. Are there opportunities along these lines for DOE/BER? The nanoscience initiative also includes some activities like this. This could be a role for BER to be playing in this initiative.

Public Comment - None

Meeting adjourned at 5:33 PM.

Ari Patrinos - Associate Director for Biological and Environmental Research

• General FY 2001 budget summary – increases primarily in the Life Sciences (the first time in a long time); slight increase for Environmental Sciences; the Medical Sciences numbers are, as usual, deceiving since they are compared to the FY 2000 base plus Congressional earmarks; we are requesting an increase in capital equipment for the Environmental Molecular Sciences Laboratory, an increase for the Climate Change Technology Initiative (CCTI) and increases in capital equipment and general plant project funding for Structural Biology

• FY 2001 "Initiatives":
  • Microbial Cell Project, $10M (+ $2.5M in BES budget request)
  • Bioengineering, $5M increase
  • Construction of the Laboratory for Comparative and Functional Genomics at ORNL ($2.5M in FY01, completion December FY04, $14.4M TEC) to replace the 50 year old facility currently in use (barely)
  • Facilities investments to more fully utilize and increase the efficiency of our user facilities; $3M in GPP for a Laboratory Office Module at the Advanced Photon Source at Argonne. This would match an investment by NIH’s National Institute of General Medical Sciences providing infrastructure support for four additional beamlines.; $4.5M in capital equipment funds for a macromolecular beamline (a DNA repiar beamline) at the Advance Light Source at Berkeley.; $3M in capital equipment for Mass Spectrometry and Nuclear Magnetic Resonance equipment at EMSL for proteomics and structural genomics.

• Microbial Cell Project – a joint BER/BES project to model and understand a complete microbial cell; this project offers many options scientifically; the request is small compared to task; there is broad support for this project from our labs and across the broader scientific community

• Bioengineering – We funded a set of pilot project in July 1999 that have made excellent progress. We are holding a contractor meeting in Albuquerque May 16-18. We have developed a strong collaboration with the NIH Bioengineering Consortium (BECON).

• Medical imaging research is focusing on the imaging of gene expression

• A top priority is the completion and implementation of the BERAC report that will arise from yesterday’s discussion. This will send a much needed powerful message to our management and to Congress about BER’s role in the next stage of biotechnology revolution. This report still needs what some have called the "10 second elevator title" or description.

• Structural genomics – There is a lot going on here at the federal and international levels. There was a meeting in Cambridge last month hosted by the Wellcome Trust. By analogy with the human genome program it was a "Bermuda-type" meeting of funding agency representatives for Structural Genomics. Structural genomics has multiple definitions and a widely varied scope and breadth. At a general level is looking to do for proteins what the genome program has done for genes and DNA. The Cambridge meeting focused on data release and data sharing policies, the development of common standards, issues of peer review, and the roles of the private sector. There will be a related meeting as part of the Global Science Forum meeting in Italy in June.

• Low Dose Radiation Research Program – We keep asking for a lot less than BERAC recommended and therefore continue to rely on Congressional to add additional funds.

• Human genome – Our working draft of the DNA sequence of human chromosomes 5, 16, and 19 is completed and was announced by the Secretary on April 13. The Secretary is visiting the Joint Genome Institute (JGI) today. We hope to have a process in place within weeks to help us identify the path forward for JGI, including, sequencing targets and links to other initiatives. The ongoing saga of public/private collaboration continues. Interaction is still a possibility.

• Restructuring of the Medical Sciences division – focus on molecular nuclear medicine and bioengineering.

• EMSL user facility – This facility continues to have challenges due to its remoteness but it has done very well in past few years in terms of the quality of science being done. We just want more.

• Natural and Accelerated Bioremediation Research (NABIR) – We have funding shortfall concerns in FY01 when we will be $4M short compared to previous years. (Over the years BER has generally grown through new initiatives rather than by adding funds to good programs.) We hope to be able to restore these funds. NABIR does have strong linkages to other programs that are growing such as the Microbial Cell Project. We now have a Field Research Center at ORNL.

• Global change program:
  • Our CCTI program is at $9M in FY01. We now have two carbon sequestration centers at $1M each focused on oceans and terrestrial carbon.
  • The US Global Change Research Program (USGCRP) has been restructured. The 10 year plan is finally being implemented. Richard Moss of Battelle PNNL is the new USGCRP director.
  • USGCRP enabling language from 1990 called for the conduct of National Assessments to periodically assess climate change in the US. The report from the recent first assessment is due out in the next few months. Lots of time and energy were spent on this and it is certainly not without controversy. I am hopeful that the report will help the dialog about action in climate change activities. National Assessment wouldn’t be nearly as good as it is if it were not for efforts in BER in modeling and assessment strategies.
  • BER’s priorities continue to be (1) global/regional climate change on decade/century scales, (2) enhanced modeling and the associated computer science especially implementing models on massively parallel computers, (3) continued emphasis on carbon cycle research, (4) expanded carbon sequestration research, (5) a strong commitment to the interagency USGCRP, and (6) intra-agency communication especially related to CCTI.

• Accelerated Climate Prediction Initiative (ACPI) – an interagency (NSF, NOAA, NASA, DOE) effort to deal with the recommendations of the National Academy to better implement computational capabilities in climate modeling. We have some new resources here and are leveraging those in the Information Technology for the 21^st Century programs. BER actually got to keep a piece of the ill-fated Scientific Simulation Initiative (SSI) in FY00 as part of our climate modeling activities.

• ARESE (ARM Enhanced Shortwave Experiment) – Two ARESE studies completed on aircraft/ground-based radiation measurement. ARESE is looking at short wave radiation above and below clouds - a major controversy in this community. This experiment has the potential to yield data to give a definitive answer. This could require a re-do of some of the assessments. We are hoping for resolution soon. An answer either way is an advance for the field.

Warren Washington - National Center for Atmospheric Research; Chair, BERAC Global Change Subcommittee

• Report of the Global Change Subcommittee

• The National Assessment has been carried out using a very uneven process due to the different responsibilities for different sectors of the country and for the different activities that are part of the assessment. This was a useful exercise for the Nation to have gone through. It is supposed to done every 5 years according to the USGCRP enabling legislation. It should be an ongoing versus an on/off process. DOE should remain committed to the National Assessment.

• ACPI – 2 new projects were funded to get the program started quickly.
  • Work with NSF and other lab partners to make climate models much more efficient on parallel computers.
  • End-to-end pilot or demonstration project. The goal is to ascertain on a smaller scale the feasibility of the central ACPI concept of running an ensemble of global climate change simulations, downscaling the results and turning them into useful data sets for regional climate impact and assessment research. This project involves a greatly expanded ocean observation system from the 1990’s. It uses climate models to make prediction through the middle/end of the next century. Water/timber/natural resources in the Pacific Northwest are being used as a test case.

• Report on facilities discussion.
  • Focus on environmental research where facilities are generally distributed. There are no separate operations and maintenance budgets. Gene Bierly is leading this effort. This needs to be about more than environmental facilities especially after yesterday’s discussion. There are issues that cut across all of BER and not just in the environmental areas. If BERAC is to expand this effort we should look to other agencies rather than across disciplines. Perhaps it can be limited to global change at this point but used later as a paradigm for other areas within BER. There is value in building on the general perception that DOE runs facilities well.

• Approval/disapproval of global change subcommittee report? Any action statements? What is implication of acceptance? Just basic guidance from a standing committee. This is not a report in response to a specific charge from SC-1. This is a subcommittee and BERAC working report to BER only. The report was approved by voice vote for transmittal to BER.

Dean Cole - BER Medical Sciences Division

• Biomedical engineering update

• 17 pilot projects funded in FY 99 with 13 continuing in FY 00 and FY 01. Total funding for these pilots is approximately $2M, $1.8M, and $1.7M in each of these fiscal years. Many of these pilots originated from Laboratory Directed Research and Development efforts.

• To date our program has seen its greatest successes in artificial retina, peripheral nerve regeneration, and breast cancer imaging research.

• Contractor’s meeting May 16-18 in Albuquerque:
  • project updates
  • symposia on artificial limbs and sight (two of the grand challenges in medicine)
  • posters on lab project funded by LDRD or by other agencies (i.e., projects not currently funded by this program)
  • program planning session
  • tour of Sandia National Laboratory robotics lab, microchemistry lab, sensor lab
  • meeting web site - http://www.orau.gov/biomed2000/.

• Redirection of our laser medicine program – This program has been a university program up to now and we are looking to develop stronger partnerships with the labs.

• DOE-NIH BECON activities – Nanotechnology symposium June 25-26, 2000 at NIH. DOE labs will be well represented. Both BER and BES will have booths at this symposium. The second round of the NIH Biotechnology Research Program grants were just funded. The national labs participated and received some funding. Mike Viola gave a BER program overview at the April BECON meeting. There will be a national lab briefing to BECON this spring primarily from the DP labs.

• MATES (Multi Agency Tissue Engineering Working group) – currently evaluating the world status of tissue engineering (through the World Technology Evaluation Center at Loyola College)

• Dick Swaja (ORNL) is on a two year assignment in Wendy Baldwin’s office working on bioengineering/BECON issues.

Prem Srivastava - BER Medical Sciences Division

• Imaging gene expression in vivo.

• The time is right –
  • current progress in genomics,
  • November 1995 conference and BERAC report on the Medical Applications Program (Genes, Molecules and Human Biology),
  • June 99 workshop

• Looking at gene expression across the range of DNA/RNA/proteins
  • advantages/ease of imaging at protein level, can image activity, protein size helpful
  • example of PET reporter genes encoding enzymes that convert positron emitting substances to sequestered/trapped (and labeled with 18-F) products
  • doesn’t let you image any gene at any time in any person – need more basic genetic approach

• Areas of emphasis in imaging gene expression
  • synthesis – production of radiolabeled oligos that are stable in vivo
  • targeting – rapid internalization and prolonged retention in target cells
  • imaging – sequence specific interactions with acceptable hybridization kinetics
  • resolution – decrease non-sequence specific effects

• Goal to study endogenous genes, transgenes, small imaging instrumentation

• Benefits – repetitive analysis within same animal, monitoring endogenous expression

Keith Hodgson - Wrap-up comments

• Thanks to Joanne Corcoran, Shirley Derflinger, David Thomassen and the BER staff for making this an interesting and productive meeting.

Public Comment - none

Next Meeting - December 2000 date to be determined

Meeting Adjourned at 11:11 AM

Dr. Eugene W. Bierly, American Geophysical Union
Dr. David R. Burgess, Boston College
Dr. Curt Civin, The Johns Hopkins Hospital
Dr. Claire M. Fraser, The Institute for Genomic Research
Dr. Raymond F. Gesteland, University of Utah
Dr. Richard E. Hallgren, American Meteorology Society
Dr. Willard W. Harrison, University of Florida
Dr. Keith O. Hodgson, Stanford University
Dr. Fern Y. Hunt, National Institute of Standards and Technology
Dr. James W. Mitchell, Lucent Technologies
Dr. Alan Rabson, National Cancer Institute
Dr. Janet L. Smith, Purdue University
Dr. Lisa Stubbs, Lawrence Livermore National Laboratory
Dr. James M. Tiedje, Michigan State University
Dr. Warren Washington, National Center for Atmospheric Research
Dr. Barbara Wold, California Institute of Technology

U.S. Department of Energy Staff

Jim Decker, Acting Director, Office of Science (SC)
Peggy Burris, SC
Ari Patrinos, Associate Director, Office of Biological and Environmental Research (OBER)/SC
David Thomassen, Designated Federal Officer, BERAC, OBER/SC
Shirley Derflinger, Designated Federal Officer, BERAC, OBER/SC
Michael Riches, OBER/SC
Marvin Frazier, OBER/SC
Joanne Corcoran, OBER/SC
Daniel Drell, OBER/SC
Arthur Katz, OBER/SC
Marvin Stodolsky, OBER/SC
Mike Teresinski, OBER/SC
Michael Viola, OBER/SC
Prem Srivastava, OBER/SC
Dean Cole, OBER/SC
Peter Kirchner, OBER/SC
Larry James, OBER/SC
Jerry Elwood, OBER/SC
Paul Bayer, OBER/SC
John Houghton, OBER/SC
Anna Palmisano, OBER/SC
Rickey Petty, OBER/SC
Thom Dunning, Jr., SC
Heather Stockwell, EH
Jeff Sherwood, PA
Carl W. Anderson, Brookhaven National Laboratory
Teresa Fryberger, Brookhaven National Laboratory
Bill Studier, Brookhaven National Laboratory
Robert Sweet, Brookhaven National Laboratory
Nora Volkow, Brookhaven National Laboratory
Creighton Wirick, Brookhaven National Laboratory
Michelle Buchanan, Oak Ridge National Laboratory
Norm Cutshall, Oak Ridge National Laboratory
Frank Harris, Oak Ridge National Laboratory
Steve Hildebrand, Oak Ridge National Laboratory
Reinhold Mann, Oak Ridge National Laboratory
Betty Mansfield, Human Genome News, Oak Ridge National Laboratory
Richard Swaja, Oak Ridge National Laboratory
Adam Arkin, Lawrence Berkeley National Laboratory
Michael Banda, Lawrence Berkeley National Laboratory
Rob Johnson, Lawrence Berkeley National Laboratory
William Dannevik, Lawrence Livermore National Laboratory
Jeff Wadsworth, Lawrence Livermore National Laboratory
Scott Cram, Los Alamos National Laboratory
Allen Hartford, Los Alamos National Laboratory
Jill Trewhella, Los Alamos National Laboratory
David Bader, Pacific Northwest National Laboratory
Jae Edmonds, Pacific Northwest National Laboratory
Noelle Metting, Pacific Northwest National Laboratory
Ellyn Murphy, Pacific Northwest National Laboratory
William Pennell, Pacific Northwest National Laboratory
Doug Ray, Pacific Northwest National Laboratory
Gerald Stokes, Pacific Northwest National Laboratory
Mike Knotek, Sandia National Laboratory
Terry Michalski, Sandia National Laboratory
Leonard Napolitano, Sandia National Laboratory

Other Federal Agency Attendees

Michael Holland, Office of Management and Budget
Elke Jordan, NIH/NHGRI
Mark Guyer, NIH/NHGRI
Charles Edmonds, NIH/NIGMS
Wendy Baldwin, NIH
Walt Shimmerling, NASA
Mary Toler, Battelle
Sylvia Jane Wolfe, DOE Oak Ridge Operations Office

Others

Laura Gerum, American Chemical Society
Margaret Lein, National Science Foundation
Lee Makowski, National Science Foundation
Jeffrey Fox, ASM News
David Smith
Ashley Houghton
Grant Houghton
Njema Frazier, House Science Committee
John Wooley, University of California, San Diego
David Galas, Keck Graduate Institute
Barbara Jasny, SCIENCE
Britt Hedman, Stanford Synchrotron Radiation Laboratory
Murray Schulman
Pamela Moore, Capital Publications
Tarun Reddy, Inside Energy