A Systems Biology Research Program at DOE

1
A Systems Biology Research
Program at DOE

• Ari Patrinos
• Office of Biological and Environmental Research
• U.S. Department of Energy

2
High Level Tank Wastes

• Single Shell Tanks at Hanford
• 149 tanks
• 65 leakers
• 35M gallons of wastes
• 190K tons of chemicals
• 132M curies of radioactivity
• 75 Sr-90, 24 Cs-137

• Issues
• Waste characterization -Contaminant transport in
  the subsurface
• Waste treatment/disposal -Characterization/monit
  oring
• Tank corrosion

3
DOE Subsurface Contamination

• Contaminated Soil/Groundwater
• 29 million m3 contaminated soil
• 36 million m3 mill tailings
• 4.7 billion m3 contaminated groundwater
• Chemicals, metals and radionuclides
• Contaminant concentrations may exceed drinking
  water standards at some sites for hundreds of
  years

4
U.S. Department of Energy
Climate Modeling
Office of Science
Increased computational capability enables
increased resolution for realistic simulations of
ocean (and atmosphere) processes.
Note the emergence of the simulated Kuroshio
current off the coast of Japan with
increased resolution.
5
(No Transcript)
6
A Diverse Technology Strategy Supports a Broad
and Competitively Balanced Portfolio (Theres No
Silver Bullet)

• Nuclear
• Fission
• Fusion
• Carbon Capture and Storage
• Advanced Transformation Systems
• Electricity
• Hydrogen
• Bio-derivative fuels

• Energy Intensity Improvements
• Industry
• Buildings
• Transportation
• Wind and Solar
• Biotechnology
• Soils
• Biomass crops
• Advanced biotechnology

7
Achieving the Reference Case Will Not Necessarily
Be Easy

• Assumed Advances In
• Fossil Fuels
• Energy intensity
• Nuclear
• Renewables

• Gap Technologies
• Carbon capture disposal
• Adv. fossil
• H2 and Adv. Transportation
• Biotechnologies
• Soils, Bioenergy, adv. Biological energy

The Gap
8
Adding to the Portfolio Will Be Essential
Teragrams of carbon
9
60
90
10
(No Transcript)
11
Sequestering carbon in terrestrial
ecosystemsMultiple benefits, but research is
needed

• Potential opportunities
• Optimize land use
• Biomass above below ground
• Soil groundwater carbon
• Multiple benefits
• Reduced erosion
• Improved soil fertility
• Increased water retention
• Research needed
• Ecosystem dynamics assessment
• Soil improvement
• Land ecosystem management
• Species selection biotechnology
• Measurement monitoring

Small changes can lead to large benefit
Photosynthesis
62 GtC/y
-60 GtC/y
Respiration
12
The Populus Tree for Carbon Sequestration
A Populus tree
Greenhouse testing
13
U.S. Department of Energy
Ocean Carbon SequestrationSouthern Ocean Iron
Enrichment Experiments (SOFeX)
Office of Science
Iron added to ocean surface (50 ppt, 100 times
background) triggered massive phytoplankton bloom
Vertical profiles of particulate organic carbon
(POC) outside (left) and inside (right) an
enriched patch
Phytoplankton from outside (left) and inside
(right) an enriched patch
14
Antarctica
Sites of ocean fertilization
New Zealand
15
Remotely Operated Vehicle (the Tiburon) used to
test the environmental impacts of deep sea CO2
sequestration
16
Deep-Sea CO2 Experiment
Closeup of CO2 release
Collecting animals with suction sampler
CO2 Treatment Station At 3,000 meters
CO2 release into Corral
Placing animals in cage
Time lapse video
Benthic organisms sensitive to pH drop that
occurs as CO2 diffuses out of initially formed
clathrate
Animal cage
Full CO2 corral
17
Biotechnology for Energy

• Terrestrial plants produce 120B metric tons of
  biomass per year
• The stored energy is 2,400 Quads (1 Quad1015BTU)
• World energy consumption for 2001 315 Quads
• In the U.S., 85 of energy needs from coal,
  natural gas and petroleum
• Carbon emissions in the U.S. 1.6B tons of C
• 25 Quads of petroleum had to be imported
• Carbon neutral energy in the U.S. only 3 of
  total
• Only 0.2 Quads of corn-derived ethanol

POTENTIAL FOR BIOMASS IS GREATER!
18
Biomass Potential

• 2B gallons of ethanol produced per year in U.S.
• 2 stage process corn carbohydrate
  carbohydrate yeast
  ethanol
• 1 bushel of corn 2.5 gallons of
  ethanol
  (one acre 125 bushels)
  Entire U.S. corn crop
  22 billion gallons of ethanol
  20 of
  U.S. automotive needs
• efficiency of conversion of sunlight to
  carbohydrate 1 efficiency of conversion of
  cornstarch to ethanol 50

Overall Efficiency 0.5
A SMALL INCREASE IN
THE EFFICIENCY CAN SIGNIFICANTLY
INCREASE THE POTENTIAL!!
19
Microbes Produce Various Fuels
H2O CO2 O2
Yeast Clostridia
Methanogens
Ethanol Hydrogen
Methane
20
A More Efficient Process for Hydrogen Production
21
Green Algae

• Sulfur deprivation induces hydrogenases
• Can a continuous process for hydrogen production
  be developed?
• For example, a reengineered hydrogenase with
  lower sensitivity to oxygen

22
GenomicsGTL A Systems Biology Research Program
From Molecules to Cells to Ecosystems
Ecosystems
Subcellular
Cellular
Identification, subcellular location, and
dynamics of molecular machines
Regulation of gene expression in individual cells
Who is expressing what, when, where, and under
what conditions? How do they work together?
23
GTL Program Goals
GTL Program Goals
Using DNA sequence and high-throughput
technologies
goal 1 Identify and characterize the molecular
machines of life goal 2 Characterize gene
regulatory networks goal 3 Characterize the
functional repertoire of complex microbial
communities in their natural environments at the
molecular level goal 4 Develop the computational
capabilities to advance understanding of complex
biological systems and predict their behavior
Systems Biology Gain a comprehensive and
predictive understanding of the dynamic,
interconnected processes underlying living systems
24
Microbes Provide Biotechnology Payoffs for the
Nation
25
so we should venture on the study of every kind
of animal without distaste for each and all will
reveal to us something natural and something
beautiful. Aristotles On the Parts of Animals
26
Tree of Life
Bacteria Archaea
Eukarya
Euryarchaeota
Methanosarcina

• Animals

• Purple bacteria

• Gram-positive

• Halophiles

• Methanobacterium

• Fungi

• Methanococcus

• Cyanobacteria

Chlamydiae
Thermoplasma
Thermococcus

• Plants

Slime molds
Crenarchaeota
Thermoproteus Pyrodictium

• Flavobacteria

Entamoebae
Ciliates

• Spirochetes

• Deinococci

Green nonsulfur bacteria

• Stramenophiles

• Thermotogales

• Trichomonads

• Aquifex

Microsporidia
Diplomonads

• Representative species completely sequenced or
  in the process of being sequenced by DOE6 Feb
  2003

27
Sequencing to Date (6/29/04)

• http//www.genomesonline.org/
• Published Complete Genomes (including 4
  chromosomes) 199
• Prokaryotic Ongoing Genomes 508
• Eukaryotic Ongoing Genomes (including 8
  chromosomes) 421
• Total 1,128
• (18 increase since 1/13/04)

28
Using the natural diversity of microbes
to find biotechnology solutions
Methane production
Ocean carbon pumping
29
7 Core GenomicsGTL Projects
Molecular Machines
Ecogenomics Genomics
Shewanella Federation
30
GenomicsGTL Center for Molecular and Cellular
Systems

• Oak Ridge National Lab Pacific Northwest
  National Lab Argonne National Lab Sandia
  National Lab U of North
  Carolina, Chapel Hill U of Utah
• Develop and use technologies needed to identify
  and characterize a complete set of microbial
  multiprotein complexes
• Focus on carbon fixing, hydrogen producing,
  organic degrading microbe R. palustris
  Shewanella
• Current goal characterize 500 complexes this
  year!

ORNL/PNNL Protein Complex Pipeline
31
Rapid Deduction of Stress Response Pathways in
Metal/ Radionuclide Reducing Bacteria

• Lawrence Berkeley National Lab Sandia
  National Lab
  Oak Ridge National Lab
  U of Washington
  U of Missouri
  Miami U
  Diversa
• Develop computational models for behavior of
  microbial gene regulatory networks in response to
  environmental conditions at DOE waste sites

Targets - Desulfovibrio, Geobacter, Shewanella

• Implemented bacterial systems biology
  pipeline controlled biomass production,
  physiologic profiling, metabolomics, imaging,
  proteomics, computational framework for
  comparative analysis of genomic and functional
  genomic data
• Rapid assessment of the effect of
  environmental conditions on cellular behavior

32
Carbon Sequestration in Synechococcus From
Molecular Machines to Hierarchical Modeling

• Sandia National Lab
  Oak Ridge National Lab
  Lawrence Berkeley National
  Lab Los Alamos
  National Lab
  U California, San Diego
  U of
  Georgia
  U of Michigan
  
  U of California, Riverside
  U of Illinois
  National
  Center for Genome Resources
• Develop experimental/computational methods to
  understand proteins, protein-protein interactions
  regulatory networks in a microbe with a
  significant role in the carbon cycle
• Developed new microarray analysis method with
  increased sensitivity and reduced noise.
• Developed microbial simulation model that will
  enable microbial biochemistry to be modeled
  together with the behavior of the molecular
  machines that carry out those reactions

33
Microbial Ecology, Proteogenomics and
Computational Optima

• Harvard U
  Massachusetts Institute of Technology Brigham and
  Womens Hospital
  Massachusetts General Hospital
• Proteins, protein-protein interactions, gene
  regulatory networks, community behavior,
  computational models of microbes important to the
  carbon cycle bioremediation.
• New proteomics method for finding untagged
  protein complexes
• Developing whole cell flux balance model of
  Procholorococcus (major ocean photosynthetic
  organism) for use in hypothesis generation about
  its natural behavior, different strains, and gene
  knockouts
• New effort on synthetic genome

Targets Prochlorococcus (carbon) Caulobacter
(bioremediation)
34
Genetic Potential of Microbial Communities
Involved in the in situ Bioremediation of Uranium

• U of Massachusetts, Amherst
  Argonne National Lab
  U of Tennessee, Memphis
  The Institute for Genomic
  Research
• Develop computational models to predict activity
  of natural communities of microbes for
  bioremediation
• Common, naturally-occurring microbes reduce U,
  Tc, Cr other metals
• Metal reduction enhanced by feeding microbes
  carbon sources
• We can stimulate growth and activity of metal
  reducing organisms in situ

Geobacter precipitating U (Lovley, U. Mass.)
35
Shewanella Federation

• Pacific Northwest National Lab Oak Ridge
  National Lab Argonne National Lab
  Biatech
  U of Southern California
  Michigan State U
  Marine Biology Lab (Mass)
  Desert Research Inst (Nevada)
• Characterize/model biology of versatile metal
  reducing microbe to understand how they sense
  respond to their environment.
• Most comprehensive study of microbe from
  proteomics to biochemistry to imaging to pathway
  modeling to bioremediation potential

36
U.S. Department of Energy
First Steps to a Synthetic Genome
Office of Science

• Institute for Biological Energy Alternatives
• Independent bioethical review
• 5,386 base pair bacteriophage synthesized in 10
  days from a pool of 42 base pair, chemically
  synthesized, gel purified oligomers
• Oligomers ligated and converted into full length
  molecules using polymerase cycling assembly
• Accuracy demonstrated by DNA sequencing and phage
  infectivity
• First step to a new field of (much larger and
  more challenging) synthetic genomes research

37
Microbial Solutions from
the Sargasso Sea?

• Institute for Biological Energy Alternatives
• Collection of 200 L samples with shotgun
  sequencing
• Preliminary analysis suggests presence of 1,600
  species

• Identification of gt1.3 million new genes
• Discovery of 800 distinct rhodopsin homologs
  (light-sensitive photoreceptors)
• Understanding the genetic and biochemical
  diversity in our oceans may lead to new methods
  for carbon sequestration or alternative energy
  production.

38
Ecogenomics A New Frontier

• lt 1 of microbes are culturable
• Many unculturables live in interdependent
  consortia of considerable diversity
• Ecogenomics applying the tools of genomics,
  proteomics, etc to ecology
• Can we recover genome-scale sequences and reveal
  metabolic capabilities?
• What is the structure of natural microbial
  populations? How do they interact? Are they
  interdependent?
• Can we harness their metabolic capabilities?

39
Ecogenomics Studying an environment with
minimal microbial complexity
Iron Mountain - microbially mediated toxic metal
discharge Superfund Site, pH lt 1

• Biofilm with a few uncultured organisms
• Identification, through genome assembly, of
  the organisms present and responsible
• Understanding of microbial interactions,
  roles and molecular mechanisms could lead to
  biochemical solutions

40
There are two kinds of scientific revolutions,
those driven by new tools and those driven by new
concepts The effect of a concept-driven
revolution is to explain old things in new ways.
The effect of a tool-driven revolution is to
discover new things that have to be explained. In
almost every branch of science, and especially in
biology and astronomy, there has been a
preponderance of tool-driven revolutions. We have
been more successful in discovering new things
than in explaining old ones. Imagined Worlds,
Cambridge, MA Harvard University Press, 1997.
pp49ff
41

• New tools for looking at molecular interactions
  and molecular machines
• Computational tools to model microbial pathways
  behavior
• New tools to study uncultured microbes, microbial
  communities microbial metabolites
• Piloting new methods for high throughput protein
  tag production

42
High-throughput Proteomics Deinococcus
radiodurans R1

• 3.1 megabase 3,116 predicted ORFs
• Large redundancy in DNA complement
• Potential relevance to bioremediation and
  understanding DNA repair
• 83 of all predicted proteins observed/
  confirmed using AMT tags

43
GTL Facilities for the Future of Science A
Twenty-Year Outlook
Cellular Activities Understanding how cells
respond to environmental cues
Cellular Components Providing the basis to study
proteins in living systems
Developing a predictive understanding of the
functions of cells and communities of cells
Understanding how molecular machines are formed
and how they function
A New Infrastructure for Biological Research
44
Discrete Automata models
Organisms
Finite element models
Evolutionary Processes
Ecosystems and Epidemiology
Organ function
Electrostatic continuum models
Cells
Cell signalling
Biology Involves Many Scales of Time and Size
DNA replication
Size Scale
Enzyme Mechanisms
Biopolymers
Ab initio Quantum Chemistry
Protein Folding
Empirical force field Molecular Dynamics
Atoms
Homology-based Protein modeling
First Principles Molecular Dynamics
Geologic Evolutionary Timescales
10-15
Timescale (seconds)
45
High-Performance Computing Roadmap for
the Genomics GTL Program
Protein machine Interactions
?
1000 TF 100 TF 10 TF 1 TF
Molecule-based cell simulation
Molecular machine classical simulation
Cell, pathway, and network simulation
Community metabolic regulatory, signaling
simulations
Constrained rigid docking
Constraint-Based Flexible Docking
Current U.S. Computing
Genome-scale protein threading
?
Comparative Genomics
Teraflops
Biological Complexity
46
GTL Experiment TemplateGenerating Petascale Data
Sets
While this example does not account for data
processing and compression it illustrates how
even simple raw data storage will quickly become
a bottleneck for biologists.
47
GTL-type Science will Require High Performance
Computing for Both Capacity and Capability
Problems
48
Biology and Computing

• Shift from qualitative, data-poor,
  experiment- driven to quantitative, data-rich
  discipline where simulations guide and interpret
  experiments
• Flood of diverse data DNA sequence, protein
  structures, imaging, etc.
• Use of first principle methods originally
  developed for chemistry

However

• Much about even the simplest of life forms is
  terra incognita
• No overarching theories to provide context for
  experimental data at the level of organisms,
  cells, etc.
• Biology defies Occams Razor

49
Computers and Biology

• Two Principal Applications
• Bioinformatics
• Biomolecular simulations
• Bioinformatics Assembly, annotation, analysis,
  comparison of DNA sequences, protein structures,
  etc.
• Biomolecular Simulations Chemical modeling
  techniques applied to biochemical phenomena
• Emerging Areas Kinetic models of
  metabolic/regulatory pathways and
  reaction-diffusion models of transport within
  and between cells

50
Bioinformatics

• Significant research challenges in sequence
  analysis and higher-level genome annotation
• Long way from reliable protein-function
  prediction from sequence
• Large data sets from emerging biotechnologies
  expression levels, structures of protein
  complexes and subcellular resolution of
  biochemicals
• Need for reliable, robust, and more efficient
  tools, e.g., for genome
  reconstruction and gene finding also protein
  comparative modeling based on distant homologies

51
Biomolecular Simulations

• Holy Grail atomic-level simulation of every
  biochemical process in an organism
• Difficulties
• large size of biomolecules
• long simulation times (for biology)
• subtle energetics of biochemical reactions
• biochemistry far from equilibrium
• Range of Modeling Tools
• One extreme is quantum mechanical calculations
  high accuracy but computationally is limited to
  small molecular systems
• The other extreme is molecular dynamics using
  simplified ball-spring force fields
  limited accuracy
• Combination of the two approaches e.g., (II)
  providing data on large-scale conformational
  changes in enzymes and (I) predicting the effects
  of catalytic activity of the enzyme

52
http//DOEGenomesToLife.org
53
To give away money is an easy matter and in any
mans power. But to decide to whom to give it,
and how much and when, and for what purpose and
how, is neither in every mans power nor an easy
matter. Hence, it is that such excellence is
rare, praiseworthy, and noble. Aristotle