Computational biology' Detecting compound action and finding disease genes'

1
(No Transcript)
2
hedule for the lectures
 
3
hedule for the lectures
 
4
hedule for the lectures
 
5
(No Transcript)
6
hedule for the lectures
 
7
Overview

• Systems Biology some current definitions and
  perceptions of the area
• Some historical scientific roots
• Why systems biology now ?
• Other life science areas using system approaches
• A systematic account of systems biology
• Content of the course
• What is not covered in the course
• Some key issues in systems biology

8
SYSTEMS BIOLOGY - Hype ?
9
Overview

• Systems Biology some current definitions and
  perceptions of the area
• Some historical scientific roots
• Why systems biology now ?
• Other life science areas using system approaches
• A systematic account of systems biology
• Content of the course
• What is not covered in the course
• Some key issues in systems biology

10
1. Systems Biology is all about networks of -
genes - proteins - metabolites - cells -
internet - air ports - actors - spread of
diseases And the interactions
11
Compound/Drug/Disease
Molecular disease maps
12
2. SYSTEMS BIOLOGY - too large for academia ?
13
3. SYSTEMS BIOLOGY - old school ?
Decoding the logic of life
14
4. The In silico dream
hedule for the lectures
 
15
Overview

• Systems Biology some current definitions and
  perceptions of the area
• Some historical scientific roots
• Why systems biology now ?
• Other life science areas using system approaches
• A systematic account of systems biology
• Content of the course
• What is not covered in the course
• Some key issues in systems biology

16
Historical Scientific Roots

• Physiology

17
Historical Scientific Roots

• Physiology
• Applied Mathematics local models

18
Historical Scientific Roots

• Physiology
• Applied Mathematics local models
• Physics biophysics detailed modeling

19
Historical Scientific Roots

• Physiology
• Applied Mathematics local models
• Physics biophysics detailed modeling
• Physics biophysics measurement devices
• Physics statistical mechanics
• Numerical analysis stochastic, ODE, PDE
  solvers - simulations
• Cybernetics

20
Historical Scientific Roots

• Physiology
• Applied Mathematics local models
• Physics biophysics detailed modeling
• Physics biophysics measurement devices
• Physics statistical mechanics
• Numerical analysis stochastic, ODE, PDE
  solvers - simulations
• Cybernetics
• Systems theory

21
Historical Scientific Roots

• Physiology
• Applied Mathematics local models
• Physics biophysics detailed modeling
• Physics biophysics measurement devices
• Physics statistical mechanics
• Numerical analysis stochastic, ODE, PDE
  solvers - simulations
• Cybernetics
• Systems theory
• Complex networks santa fe

22
Historical Scientific Roots

• Physiology
• Applied Mathematics local models
• Physics biophysics detailed modeling
• Physics biophysics measurement devices
• Physics statistical mechanics
• Numerical analysis stochastic, ODE, PDE
  solvers - simulations
• Cybernetics
• Systems theory
• Complex networks santa fe
• Applied Mathematics axiomatic, Turing like
  approaches natural computation
• Applied Mathematics complexity, chaos,
  non-linear dynamics
• Theoretical Ecology dynamics, populations
  game theory

23
Historical Scientific Roots

• Physiology
• Applied Mathematics local models
• Physics biophysics detailed modeling
• Physics biophysics measurement devices
• Physics statistical mechanics
• Numerical analysis stochastic, ODE, PDE
  solvers - simulations
• Cybernetics
• Systems theory
• Complex networks santa fe
• Applied Mathematics axiomatic, Turing like
  approaches natural computation
• Applied Mathematics complexity, chaos,
  non-linear dynamics
• Theoretical Ecology dynamics, populations
  game theory
• Theoretical Biology understanding principles
  of life Schrödinger 1949

24
Historical Scientific Roots

• Physiology
• Applied Mathematics local models
• Physics biophysics detailed modeling
• Physics biophysics measurement devices
• Physics statistical mechanics
• Numerical analysis stochastic, ODE, PDE
  solvers - simulations
• Cybernetics
• Systems theory
• Complex networks santa fe
• Applied Mathematics axiomatic, Turing like
  approaches natural computation
• Applied Mathematics complexity, chaos,
  non-linear dynamics
• Theoretical Ecology dynamics, populations
  game theory
• Theoretical Biology understanding principles
  of life Schrödinger 1949
• History of networks Euler, Erdos, Barabasi

25
Historical Scientific Roots

• Physiology
• Applied Mathematics local models
• Physics biophysics detailed modeling
• Physics biophysics measurement devices
• Physics statistical mechanics
• Numerical analysis stochastic, ODE, PDE
  solvers - simulations
• Cybernetics
• Systems theory
• Complex networks santa fe
• Applied Mathematics axiomatic, Turing like
  approaches natural computation
• Applied Mathematics complexity, chaos,
  non-linear dynamics
• Theoretical Ecology dynamics, populations
  game theory
• Theoretical Biology understanding principles
  of life Schrödinger 1949
• History of networks Euler, Erdos, Barabasi
• Control Theory black box system
  identification of linear systems
• Statistics Machine Learning - handling
  large-scale data-sets
• Computer Science databases, ontology,
  knowledge representation and integration
• Computer Science efficient algorithms
  visulization

26
Overview

• Systems Biology some current definitions and
  perceptions of the area
• Some historical scientific roots
• Why systems biology now ?
• Other life science areas using system approaches
• A systematic account of systems biology
• Content of the course
• What is not covered in the course
• Some key issues in systems biology

27
Why systems biology now ?

• technology driven, data explosion
• need to understand biology, patterns
  correlations are not enough

28
Why systems biology now ?

• technology driven, data explosion
• need to understand biology, patterns
  correlations are not enough
• Local molecular biology not sufficient for
  understanding biological complexity

29
hedule for the lectures
 
30
Why systems biology now ?

• technology driven, data explosion
• need to understand biology, patterns
  correlations are not enough
• Local molecular biology not sufficient for
  understanding biological complexity
• Genomics has not delivered drugs as expected
• networks everywhere
• complicated systems require computational tools
  for understanding

31
(No Transcript)
32
hedule for the lectures
 
33
Overview

• Systems Biology some current definitions and
  perceptions of the area
• Some historical scientific roots
• Why systems biology now ?
• Other life science areas using system approaches
• A systematic account of systems biology
• Content of the course
• What is not covered in the course
• Some key issues in systems biology

34
Neuroscience
35
(No Transcript)
36
(No Transcript)
37
Challenges Lessons

• Data integration not solved

38
Challenges Lessons

• Data integration not solved
• Validation of models problematic without
  experimental prediction and testing
• Not comprehensive models exclude parts of
  system (pro-con)
• Simplified models have provided insights, complex
  models less
• Important to study relevant systems, i.e simple
  organisms does not translate into human
  behaviour. Back to cortex.
• Model similarity neurons (continous-discrete)
  to genes proteins.
• Well understood basic models of node (cell)
  dynamics in neuroscience but less so for gene
  regulatory systems
• Network dynamics more explored in neuroscience
  immunology but more data on network structure for
  genomic/protein/metabolic systems
• Parallel measurement technologies are lagging
  behind in neuroscience
• Neuroscience theory issues on representation
  etc due to cognitive domain. Other organs/systems
  viewed as advanced control systems

39
Overview

• Systems Biology some current definitions and
  perceptions of the area
• Some historical scientific roots
• Why systems biology now ?
• Other life science areas using system approaches
• A systematic account of systems biology
• Content of the course
• What is not covered in the course
• Some key issues in systems biology

40
Systems Biology

• Systems Biology is Experimental Computational
• What is a system is context/problem dependent.
• Experiments range from simple model system to
  humans (manipulations vs relevance)
• To understand systems there are four levels of
  analysis

41
Systems Biology

• Data administration representation (a) in
  house experiements, (b) other experimental data,
  (c) prior knowledge and other databases.

42
Systems Biology

• Data administration representation (a) in
  house experiements, (b) other experimental data,
  (c) prior knowledge and other databases.
• Detecting statistical significant features and
  patterns in data (nodes of interest)

Trial and error Look for what you
expect Unsupervised techniques Rigorous
statistics Machine learning (kernels, svm)
43
Systems Biology

• Data administration representation (a) in
  house experiements, (b) other experimental data,
  (c) prior knowledge and other databases.
• Detecting statistical significant features and
  patterns in data (nodes of interest)
• Underlying biology that generates the observed
  patterns Identify edges in the network under
  study.

44
(3a) Different types of edges - semantics

• A network is built from nodes
• and edges connecting them.
• Edges can be
• directed (hyperlinks, gene regulation)
• or
• undirected (friendships, streets)
• Edges can have weights (friendship) or
• other properties

45
(3b) Different types of graphs
a directed tree
a directed acyclic graph
a complete graph
46
(3c) Different types of biological networks
GENOME
PROTEOME
METABOLOME
47
Two strategies to identify edges

• Collect prior edges and project experimental data
  on top of the scaffold.

hedule for the lectures
 
48
Two strategies to identify edges

• Collect prior edges and project experimental data
  on top of the scaffold.
• Infer edges by combining network identification
  algorithms high-throughput data

hedule for the lectures

• Underlying computational model with parameters at
  some level of resolution
• Experimental data
• Fitting procedure to identify model parameters
  identify edges

 
49
Scope of the challenge
Number of components
hedule for the lectures
Number of combinations
Time-scales
 
50
Systems Biology

• Data administration representation (a) in
  house experiements, (b) other experimental data,
  (c) prior knowledge and other databases.
• Detecting statistical significant features and
  patterns in data (nodes of interest)
• Underlying biology that generates the observed
  patterns Identify edges in the network under
  study.
• Dynamical modeling of the system

51
Why dynamical modeling systems are complex and
involve time
hedule for the lectures
 
52
Why dynamical modeling systems are complex and
involve time
hedule for the lectures
 
53
Why dynamical modeling exhaustive simulations
vs insight using non-linear dynamics
hedule for the lectures
 
54
hedule for the lectures
 
55
Overview

• Systems Biology some current definitions and
  perceptions of the area
• Some historical scientific roots
• Why systems biology now ?
• Other life science areas using system approaches
• A systematic account of systems biology
• Content of the course
• What is not covered in the course
• Some key issues in systems biology

56

• YELLOW Network identification level 3
• BLUE Computational modeling ( experiments)
  level 4
• RED Integrative physiological/medical
  approaches mix of level 1, 2, 3

57
Systems Biology

• Data administration representation (a) in
  house experiements, (b) other experimental data,
  (c) prior knowledge and other databases.
• Detecting statistical significant features and
  patterns in data (nodes of interest)
• Underlying biology that generates the observed
  patterns Identify edges in the network under
  study. EA directed/undirected/motifs networks
  JP Bayesian network inference MH network
  inference usign ODE and regression
• Dynamical modeling of the system EA
  Phage/lambda modeling ME regulation E-Coli OW
  non-linear modeling, small circuits, pathway
  modeling HS control theory analysis

HS systems biology software Integrative
applied approaches ME E Coli JB
Cardiovascular IE Cancer KT Kidney
Physiology
58
Overview

• Systems Biology some current definitions and
  perceptions of the area
• Some historical scientific roots
• Why systems biology now ?
• Other life science areas using system approaches
• A systematic account of systems biology
• Content of the course
• What is not covered in the course
• Some key issues in systems biology

59
What is not covered in the course

• measurement/high-throughput technology
• metabolic networks and flux analysis
• Edge libraries computational prediction
  methods for edges (binding etc)
• Data standards, administration knowledge
  representation
• Statistics and feature detection

60
What is not covered in the course

• measurement/high-throughput technology
• metabolic networks and flux analysis
• Edge libraries computational prediction
  methods for edges (binding etc)
• Data standards, administration knowledge
  representation
• Statistics and feature detection
• Large scale dynamical modeling of organs

61
What is not covered in the course

• measurement/high-throughput technology
• metabolic networks and flux analysis
• Edge libraries computational prediction
  methods for edges (binding etc)
• Data standards, administration knowledge
  representation
• Statistics and feature detection
• Large scale dynamical modeling of organs
• Drug development
• Synthetic biology

62
What is not covered in the course
63
Overview

• Systems Biology some current definitions and
  perceptions of the area
• Some historical scientific roots
• Why systems biology now ?
• Other life science areas using system approaches
• A systematic account of systems biology
• Content of the course
• What is not covered in the course
• Some key issues in systems biology

64
Some key challenges problems in systems biology

• Importance of low level problems (data-admin,
  signal-noise, scripting)
• Standardization (platforms, analysis, to ensure
  reproducible results)

65
Two cultures biology versus rigorous controlled
vocabulary
hedule for the lectures
 
66
Some key challenges problems in systems biology

• Importance of low level problems (data-admin,
  signal-noise, scripting)
• Standardization (platforms, analysis, to ensure
  reproducable results)
• Integration of different data-types

hedule for the lectures
 
67
hedule for the lectures
 
68
Some key challenges problems in systems biology

• Importance of low level problems (data-admin,
  signal-noise, scripting)
• Standardization (platforms, analysis, to ensure
  reproducable results)
• Integration of different data-types
• Omics approach vs a problem-oriented approach
• Bottom-up vs top-down
• Relevance of system under study vs possibility to
  manipulate the system

hedule for the lectures
 
69
Some key challenges problems in systems biology

• Importance of low level problems (data-admin,
  signal-noise, scripting)
• Standardization (platforms, analysis, to ensure
  reproducable results)
• Integration of different data-types
• Omics approach vs a problem-oriented approach
• Bottom-up vs top-down
• Relevance of system under study vs possibility to
  manipulate the system
• Level of coarse graining for modelling system of
  interest
• Stochastic vs Deterministic model (ODE, PDE)

hedule for the lectures
 
70

• Level of coarse graining for modeling system of
  interest
• Stochastic vs Deterministic model (ODE, PDE)

hedule for the lectures
 
71
Some key challenges problems in systems biology

• Importance of low level problems (data-admin,
  signal-noise, scripting)
• Standardization (platforms, analysis, to ensure
  reproducable results)
• Integration of different data-types
• Omics approach vs a problem-oriented approach
• Bottom-up vs top-down
• Relevance of system under study vs possibility to
  manipulate the system
• Level of coarse graining for modeling system of
  interest
• Stochastic vs Deterministic model (ODE, PDE)
• Validation.

hedule for the lectures
 
72
Validation.
Statistics Prior knowledge Exp prediction
test
hedule for the lectures
Features (nodes, patterns) Network Computati
onal Model
 
73
Validation.
Statistics Prior knowledge Exp prediction
test
hedule for the lectures
X
X
Features (nodes, patterns) Network Computati
onal Model
X
X
 
74
hedule for the lectures
 
75
A different kind of validation
Synthetic Biology refers to A) the design and
construction of new biological parts, devices,
and systems. B) the re-design of existing,
natural biological systems for useful purposes.

• CODON DEVICES

76
Summary

• Systems Biology some current definitions and
  perceptions of the area
• Some historical scientific roots
• Why systems biology now ?
• Other life science areas using system approaches
• A systematic account of systems biology
• Content of the course
• What is not covered in the course
• Some key issues in systems biology