the canadian bioinformatics workshops cbw series bega - PowerPoint PPT Presentation

Loading...

PPT – the canadian bioinformatics workshops cbw series bega PowerPoint presentation | free to view - id: 13775-YzJkY



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

the canadian bioinformatics workshops cbw series bega

Description:

the canadian bioinformatics workshops cbw series began offering one and two week short courses in bioinformatics, genomics and proteomics in 1999, in response to an identified need for a skilled bioinformatics workforce in canada. in partnership with the canadian genetics diseases network and human resources development canada, and under the scientific direction of director, francis ouellette, the cbw series was established. for eight years, the series offered short courses in bioinformatics, genomics and proteomics in various cities across canada. taught by top faculty from canada and the us, the courses offered small classes and handson instruction. in the fall of 2007, at the third mentor network meeting, a new format and series of exciting workshops were designed. it was recognized that with the introduction of new technologies and scientific approaches to research, having the computational biology capacity and skill to deal with this new data has become an even greater asset. this new series of workshops focuses on training the experts and users of these advanced technologies on the latest approaches being used in computational biology to deal with the new data. the cbw plans to offer these workshops in the summer of 2008 and 2009. – PowerPoint PPT presentation

Number of Views:220
Avg rating:3.0/5.0
Slides: 55
Provided by: rmaha
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: the canadian bioinformatics workshops cbw series bega


1
Canadian Bioinformatics Workshops
  • www.bioinformatics.ca

2
(No Transcript)
3
Constraint-based Models of Metabolism
  • R. Mahadevan,
  • Assistant Professor, Department of Chemical
    Engineering and Applied Chemistry, Institute of
    Biomaterials and Biomedical Engineering
  • University of Toronto

4
Outline
  • Introduction/ Motivation for Modeling
  • Constraint-based Modeling
  • Applications of CBM

5
Driving Forces for in silico Models
Systems Biology
6
Systems Biology Approach
  • study of how the parts work together to form a
    functioning biological system (Church
    coworkers, 2003)

7
Growth in Systems Biology Research
8
Era of In silico models
  • Over 353 micro-organisms sequenced (TIGRs CMR
    database)
  • Genome sequence alone has provided limited
    information on the phenotype (e.g., genes for O2
    utilization, glucose uptake )
  • Abundance of this sequence information motivates
    the in silico modeling era
  • Genome-scale models available for a variety of
    microbes relevant to biotech, pharma industry

9
Applications of Models
  • Discovery through combined computation and
    experimentation
  • Functional genomics
  • Integrative data analysis
  • Physiology Analysis (Hypothesis generation)
  • Design
  • Metabolic Engineering
  • Protein Production
  • Bioprocess Optimization

10
Methods to Interrogate Models
Price, et al Nat Rev Microbiol. Oct 2004
  • Sophisticated methods to evaluate systemic
    properties for Discovery and Design based on the
    Network

11
2D Annotation Networks
  • Metabolic

Regulatory
Signaling
1) Network reconstruction first step in the
analysis of network function 2) All network
interactions are biochemical interactions
covalent bonds or weak interactions
Protein-DNA
Protein-Protein
12
Network Interactions
  • Interactions have specific properties
  • Chemical Interactions (Covalent Bonds) e.g
    metabolic networks
  • Stoichiometry (mostly invariant)
  • Kinetic rates (condition dependent)
  • Thermodynamics (determines directionality)
  • Hard links (easier to determine)
  • Physical Interactions (Hydrogen bonding)
    protein-protein networks
  • Soft links that can form readily
  • Can be condition dependent and difficult to
    reconstruct with confidence

13
Network Interactions Continued
  • Estimate of Network Components Possible (e.g. no.
    of genes, proteins, metabolites (?))
  • Interactions can be combinatorial
  • e.g interaction network between 1000 proteins has
    close to half a million elements ! (1e6-1000)/2
    (n2-n)/2 interactions for n proteins ?
  • Not all interactions occur (Selection guided by
    evolution to accomplish spec. mechanisms!)
  • Required for complexity in phenotypes (30,000
    genes in humans ?)

14
Constraints in Biology
  • Physico-chemical constraints
  • Enzyme capacity (Vmax)
  • Conservation (mass, energy, charge )
  • Environment
  • Substrate availability (donor/acceptors)
  • pH, temperature, pressure etc...
  • Regulatory constraints
  • Availability of proteins
  • Spatial constraints
  • DNA arrangement
  • Local concentrations

15
Biological Networks Summary
  • Complex networks (with biochemical and physical
    interactions) characterized by uncertainty
  • Combinatorial interactions that can vary in time
  • Even biological networks have to satisfy specific
    constraints
  • Networks can be modular with specific functional
    attributes

16
2D Annotation Networks
  • Metabolic

Regulatory
Signaling
Focus on Metabolic Networks due to ease of their
Reconstruction
Protein-DNA
Protein-Protein
17
Model-Building Process (e.g.,Metabolism)
  • Determine network components
  • Define links between components
  • Quantitative calculations for modeling

18
Process of Metabolic Reconstruction(Covert et
al., 2001)
Selecting Pathway for Construction
Identifying all genes in genome for selected
pathways
gene/transcript
polypeptide
Reviewing sequence annotation (focused BLAST
searching in-house)
protein
reaction
Cross Validation to external references (NCBI,
CMR and JGI)
Associating genes-proteins-reactions
Complete Network Reconstruction
19
Constraint-based Modeling of Metabolism
Parameters
  • ATP maintenance parameters
  • Growth and non-growth associated ATP terms
  • Calculated by regression with experimental data
  • Biomass composition
  • Determined from data
  • Metabolite transport rates
  • Limiting substrate uptake rates (inputs to the
    model)
  • Obtained from measurements

20
Determining ATP Maintenance Parameters
Experimental data
Growth Yield (Biomass/Substrate)
Growth Rate
  • Determine the maintenance parameters based on
    data
  • Calculate from biomass protein composition, ATP
    required for protein synthesis and other cellular
    process
  • Incorporate in the biomass reaction

21
Representing Biomass Composition in the Model
Uptake Rates Acetate, Fumarate, Ammonium, etc.
Biomass Reaction
w biomass Protein content 46 RNA
content 10 DNA content 4 Carbohydrate
content 15 Lipid content 15 Other 10
22
Mathematical Representation of a Biochemical
Network
Stoichiometric Matrix
reaction
metabolite
S
S matrix concise and systematic representation
of the all the reactions, metabolites and their
interactions
23
Obtaining the Stoichiometric Matrix (S)
  • S obtained from the list of reactions
    reconstructed from the combination of genomics,
    biochemistry and physiology

chemical reaction
Reaction list
vi
A -a B 0 C -c D 0 E e F 0 G 0 H h
compounds
vi reaction a,b,e,h stoichiometric
coefficients (invariant) A,C, E, H
compounds/metabolites
24
Representation of the 2D Annotation Metabolic
Networks
  • Mathematical representation of the components and
    their interactions required for calculations
  • Define component concentration vector
  • x x1,x2,,xm m number of metabolites
  • Define reaction flux vector
  • v v1,v2,,vn n number of reactions
  • Typically, mltn for metabolic networks
  • Underdetermined systems with several degrees of
    freedom

25
S links metabolites and reactions
S
ST
S describes the interactions among metabolites
(substrates, products) Columns describe the
reaction stoichiometry Rows define connectivity
26
Sparsity
S is inherently sparse Example Geobacter
sulfurreducens S has 2655 nonzero out of 3e5
elements (.8 ) Some highly connected
metabolites present
Metabolites
Reactions
27
Dynamic Mass Balance
  • Mass balance around a metabolite, say pyruvate
  • Generalizing
  • S matrix transforms the reaction flux vector v
    v1,v2,,vn to the time derivatives of
    concentration

Rate of accumulation of pyr Production rate-
Consumption rate
28
Calculating Phenotypes Using a Constraint-based
Approach
Growth /Biomass Composition
Physicochemical Constraints
Convexity
29
Objective Function
  • Typical objective function is growth rate
    maximization
  • Uptake rates are also required and are usually
    specified
  • Therefore, maximal yield solution is found
  • However, this holds true only under optimal
    condition
  • Several conditions where this objective function
    fails
  • No clear alternative proposed

30
Cellular Objective of Optimality
31
Evolution of deletion mutants (Fong Palsson,
2004)
  • Knockout strains initially exhibit poor growth
  • Eventually evolve to predicted growth rates

32
Robustness Analysis to Gene Deletions and Enzyme
Defects
Biological Significance The impairment of an
enzyme can have a system wide effect and affect
the optimal growth rate achievable by an organism.
Example Fluxes in E. coli have been analyzed to
study how a continuous impairment of the enzyme
will affect the predicted optimal growth rate.
Mathematics
References Edwards, J.S., and Palsson, B.Ø.,
"Robustness Analysis of the Escherichia coli
Metabolic Network", Biotechnol Prog., 16
927-939, (2000).
33
Phenotypic Phase Planes
Biological Significance Can determine what the
optimal nutrient uptake rates to allow for
maximal biomass production (Line of Optimality)
and what uptake rates are not feasible.
2.4
Line of Optimality
0.4
Oxygen Uptake Rate
Phase Plane
Isoclines
Mathematics Shadow prices from the dual solution
are calculated for different uptake rates.
Shadow prices are constant within a region,
changes in shadow prices delineate the different
regions.
Carbon Uptake Rate
Key References Edwards, J.S., Ibarra, R.U., and
Palsson, B.Ø., "In silico predictions of
Escherichi coli metabolic capabilities are
consistent with experimental data", Nature
Biotechnology 19 125-130(2001). Edwards, J.S.,
Ramakrishna R., Palsson, B.Ø., Characterizing
the metabolic phenotype A phenotype phase plane
analysis",Biotechnology and Bioengineering,
77(1) pp. 27-36 (2002). Schilling,C.H.,
Edwards, J.S., Letscher, D.L., and Palsson, B.Ø.,
"Combining pathway analysis with flux balance
analysis for the comprehensive study of metabolic
systems", Biotechnology and Bioengineering 71
286-306 (2001). Ibarra, R.U., Edwards, J.S., and
Palsson, B.Ø. "Escherichia coli K-12 undergoes
adaptive evolution to achieve in silico predicted
optimal growth," Nature, 420 pp. 186-189 (2002).
34
Constraint-based Modeling Approach Issues
  • Linear programming with growth rate maximization
    generates flux distribution (Max vgro s.t. Sv
    0)
  • Need to be aware that multiple optimal flux
    distributions can exist for the same
    environmental conditions
  • Alternate Optima (Degenerate solutions) can be
    present
  • Same objective function value, different solution
    (flux distribution)

35
Flux Variability Analysis(Mahadevan and
Schilling, 2003, Metabolic Engineering)
  • MILP based algorithm exists to identify all
    alternate optimal solutions (Lee et al., 2000)
  • Can be intractable at the genome scale
  • Need a computationally efficient and practical
    approach
  • Flux Variability Analysis
  • Objective function value obtained in the first
    trial incorporated as additional equality
    constraint
  • Maximization and Minimization of individual
    fluxes specified as the objective function
  • LP solved for all the minimization and
    maximization of all fluxes

36
Alternate Optimal Flux Distribution
Characterization
37
Condition Dependent Flux Variation
  • Reactions associated with formate secretion show
    variation for growth on glucose
  • Range of flux variation condition dependent

38
Equivalent Reaction Sets
  • Alternate optimal solutions due to the presence
    of equivalent reaction sets
  • Modified extreme pathway algorithm to determine
    all equivalent reaction sets for growth on glucose

B
A
Loop
B
C
A
C
B
A
Equivalent Reaction Sets
C
39
Equivalent Reaction Set Example for Growth on
Glucose
  • 50 equivalent reaction sets identified
  • Equivalent reaction sets primarily in the
    nucleotide metabolism, amino acid metabolism
  • Example in Ribonucleotide reductase reactions
  • Set of 4 reactions stoichiometrically equivalent
    to the single reaction
  • Net reaction gtp trdrd -gt dgtp trdox
  • Condition dependent realization of equivalent
    reaction sets leads to alternate optima

40
Application to Physiology
  • Analysis of Geobacter Metabolism

41
Geobacter sulfurreducens
  • G. sulfurreducens
  • Important member of the dissimilatory metal
    reducing bacteria (Geobacteraceae) (Caccavo et
    al., 1994)
  • Strictly anaerobic bacteria
  • Grows with acetate as the electron donor and
    metals (Fe), fumarate as electron acceptors
  • Metabolism not well characterized (Galushko and
    Schink, 2000)
  • Genome sequencing completed at The Institute of
    Genomic Research (TIGR)
  • Applications in bioremediation and bioenergy
    generation

42
G. sulfurreducens Applications
  • Bioremediation
  • Can reduce toxic soluble metals Co(III), Tc(VII),
    U(VI) to insoluble forms
  • Clean-up of radioactive/contaminated sites

43
G. sulfurreducens Applications
  • Bioenergy generation
  • Can respire on acetate and donate electrons to
    electrode
  • Has been shown to recover 95 of electrons in
    acetate (Bond and Lovley, 2003)
  • Microbial fuel cell applications

Acetate
CO2
e-
Anode
Motivates need for improved understanding of
metabolism
44
Metabolic Network of G. sulfurreducens
Total Number of Genes 3466 Included Genes 588
(17 ) Percentage of the annotated genome (29)
Total Number of Model Reactions 522 Total
Number of Metabolites 541
45
Metabolic Modeling Results
  • General metabolic capabilities
  • Significance of pyruvate ferredoxin
    oxidoreductase for enhanced biomass yields during
    growth with acetate
  • Impact of extracellular electron acceptors
  • Differences in energy generation attributed to
    proton balancing
  • Prospective studies
  • Analysis of the energetics of menaquinone
    secretion
  • Explanation for dominance of G. sulfurreducens in
    the environment

46
Comparison with E. coli Amino Acid Synthesis
  • Genome-scale metabolic models used to compare the
    capabilities of the networks of E. coli G.
    sulfurreducens (Reed et al., 2003)
  • Ability to synthesize amino acids during acetate
    oxidation with Fe(III) as electron acceptor,
    analyzed
  • G. sulfurreducens network more efficient than E.
    coli for synthesizing amino acids

One letter code used to represent AAs on the plot
(Aalanine, etc)
47
Simulation Studies Analysis of Proton
Translocation Stoichiometry
  • Electron transport chain Key energy generating
    step
  • Transfer of electrons from a donor to an acceptor
  • Accompanied by translocation of protons across
    the membrane

ATP synthesis driven by H gradient Maintaining
the H gradient is critical
Image from Biology A guide to the natural
world David Krogh, 2002
48
Proton Translocation Stoichiometry Issues
  • Number of protons translocated per electron
    (H/e- ratio) is a critical parameter
  • H/e- ratio depends on available energy between
    the donor/acceptor (e.g, NADH/O2 ,NADH/Fumarate,
    NADH/Fe(III) )
  • Thermodynamic consideration leads to H/e- ratio
    of 1 for NADH/Fumarate (Kroger et al., 2002)
  • Experimental evidence indicates that biomass
    yield per mole of acetate with fumarate is three
    times the yield on Fe(III)

Biomass yield (gdw/mmol Ac)
49
Proton Translocation under Fe(III), Fumarate
Reduction
Acetate Fumarate
Net Rxn Ac- H Fum? CO2 Succ
Acetate Fe(III)
Net Rxn Ac-Fe(III)? CO2 Fe(II) H
Fumarate reduction Net consumption of cytosolic
protons Fe(III) reduction Net production of
cytosolic protons Electron transport chain
results in excess cytosolic proton production
during Fe(III) reduction
50
Model-based Analysis of Yield Differences
  • Excess cytosolic protons formed during Fe(III)
    reduction
  • Energy (ATP) required to pump excess protons
  • Energetics of proton pumping leads to decreased
    biomass yields
  • Model-based analysis explained observed yield
    differences
  • Key physiological insight related to proton
    generation

(Model)
(Model)
51
Need for Enhanced Power Generation Rates
  • G. sulfurreducens can respire on acetate and
    donate electrons to electrode (Bond Lovley,
    2003)
  • Low power generation rates (0.001 mW/cm2 ) even
    though high efficiency
  • Clear requirement for enhancing the rate of
    power generation through
  • strain engineering
  • electrode optimization
  • protein engineering

Low-powered devices
Cathode
52
Engineering Geobacter for Enhanced Electricity
Generation Capability
Goal Increase electricity (i.e. current)
generation capability of Geobacter 1) modify
native metabolism to increase electron transfer
rate/respiration
Cell Membrane
Acetate
e-
Cytoplasm
Electrons (e-)
Anode
Cathode
53
Model Predictions Creating Futile Cycles
  • Simulations indicate that increased ATP
    drain/maintenance leads to increased electron
    transport flux
  • ? ATPdrain ? e- transport, ? ?

54
Engineering Futile Cycle Results
  • With futile cycle
  • Geobacter (pCDatpAGD)
  • acetate fumarate (e- acceptor)
  • Without futile cycle (control)
  • Geobacter (pCD)
  • acetate fumarate (e- acceptor)
  • Rate of fumarate reduction
  • No induction 0.88 mM Succ/OD/h
  • 1mM IPTG 1.13 mM Succ/OD/h
  • Specific Level of Soluble Fe(III) reduction
  • No induction 1.19x10-7 mM FeII/cell
  • 1mM IPTG 2.4x10-7 mM FeII/cell
  • 2x ? Respiration Rate on Fe(III) due to Futile
    Cycle

Mounir Izallalen from Lovley group at UMass
About PowerShow.com