What is Systems Biology? Where does it come from? What are the challenges ahead? Nicolas Le Novère, Babraham Institute, EMBL-EBI

Transcription

1 What is Systems Biology? Where does it come from? What are the challenges ahead? Nicolas Le Novère, Babraham Institute, EMBL-EBI

2

3

4 What did electrical engineering bring? O F D O R T V A O W L E E H K R H T L A M O C N U F

5 What will biological engineering bring? FOOD HEALTH (drug production regenerative medicine) ENVIRONMENT (biofuels decontamination) (crop) SECURITY (biosensors)

6 Why using mathematical models? Describe 1917

7 Why using mathematical models? Describe 1917 Explain 1952

8 Why using mathematical models? Describe 1917 Explain 1952 Predict 2000

9 What is a mathematical model? Wikipedia (April 17th 2013): A mathematical model is a description of a system using mathematical concepts and language.

10 What is a mathematical model? Wikipedia (April 17th 2013): A mathematical model is a description of a system using mathematical concepts and language. variables [x] Vmax Kd EC50 length t1/2

11 What is a mathematical model? Wikipedia (April 17th 2013): A mathematical model is a description of a system using mathematical concepts and language. variables relationships [x] Vmax Kd EC50 length t1/2 If then

12 What is a mathematical model? Wikipedia (April 17th 2013): A mathematical model is a description of a system using mathematical concepts and language. variables relationships constraints [x] [x]>0 Vmax Energy conservation Kd Boundary conditions (v < upper limit) EC50 Objective functions (maximise ATP) length t1/2 If then Initial conditions

13 What is a mathematical model? Wikipedia (April 17th 2013): A mathematical model is a description of a system using mathematical concepts and language. variables relationships constraints [x] [x]>0 Vmax Energy conservation Kd Boundary conditions (v < upper limit) EC50 Objective functions (maximise ATP) length t1/2 If then Initial conditions Different types: Dynamical models, logical models, rule-based models, multi-agent models, statistical models, etc.

14 Emergence of the notion of system XXI Description of interacting components XX Description of the components of the world XIX XVIII Global Description of the world XVII classical mechanics, anatomy, physiology XVI Statistical physics, thermodynamics, quantum mechanic, biochemistry, structural biology, molecular biology Cybernetics, Information theory, telecommunications, automata, multi-agents, Systems Biology

15 Systems are formalised mid-xxth

16 Systems are formalised mid-xxth [A system consists of] a dynamic order of parts and processes standing in mutual interaction. [...] The fundamental task of biology [is] the discovery of the laws of biological systems" Ludwig von Bertalanfy, Kritische Theorie der Formbildung, 1928

17 The three paradigms of Biology Systems Biology XXI 2000 OMICS Molecular Biology XX Recombinant DNA Eccles/Katz 1950 XIX Physiology Monod/Jacob Watson/Crick Avery Hill/Meyerhoff Kossel XVIII 1900 Michaelis/Menten Pavlov/Langley XVII XVI 1850 Claude Bernard MCA/BST Hodgkin/Huxley

18 Computer simulations Vs. mathematical models

19 Computer simulations Vs. mathematical models One would like to be able to follow this more general process mathematically also. The difficulties are, however, such that one cannot hope to have any very embracing theory of such processes, beyond the statement of the equations. It might be possible, however, to treat a few particular cases in detail with the aid of a digital computer. This method has the advantage that it is not so necessary to make simplifying assumptions as it is when doing a more theoretical type of analysis.

20 Birth of Computational Systems Biology R.I.P. 30 May 2012 Cambridge

21 The Computational Systems Biology loop biological model validation simulation mathematical model parameterisation computational model

22 Towards Systems Biology XXI XX XIX 1980 XVIII XVII 1960 XVI

23 Towards Systems Biology XXI XX XIX 1980 XVIII Signalling pathways Gene regulatory networks Cell cycle Goldbeter/Koshland Covalent cascades Metabolic networks XVII 1960 XVI Denis Noble Heart pacemaker Hodgkin/Huxley models

24 Towards Systems Biology Organ level XXI XX XIX 1980 XVIII Tissue models Signalling pathways Gene regulatory networks Purkinje neuron Cell cycle Goldbeter/Koshland Covalent cascades Metabolic networks Simple circuits Complex neurons XVII 1960 Denis Noble Heart pacemaker Rall's cable approximation neurobiology XVI Hodgkin/Huxley models

25 60s and 70s Mihajlo Mesarovic: 1966 Symposium general systems theory and biology Stuart Kaufmann, Rene Thomas: boolean networks for gene regulation Henri Kacser: Metabolic control analysis, Michel Savageau: Biochemical Systems Theory

26 Towards Systems Biology Organ level SBML XXI XX XIX 1980 XVIII Tissue models Signalling pathways Gene regulatory networks Purkinje neuron Cell cycle Goldbeter/Koshland Covalent cascades Metabolic networks Simple circuits Complex neurons XVII 1960 Stochastic algorithms MCA/BST Boolean networks Denis Noble Heart pacemaker Rall's cable approximation neurobiology XVI User-friendly simulators Multi-agent systems Hodgkin/Huxley models methods

27 80s: The reign of Molecular Biology

28 90s: maturation of the community Publication of modelling work in high visibility journals, e.g.: Tyson. modeling the cell-division cycle - cdc2 and cyclin interactions. PNAS 1991; McAdams and Shapiro. Circuit simulation of genetic networks. Science 1995; Barkai and Leibler. Robustness in simple biochemical networks. Nature 1997; Neuman et al. Hepatitis C viral dynamics in vivo and the antiviral efficacy of interferon-alpha therapy. Science 1998; Yue et al. Genomic cis-regulatory logic: Experimental and computational analysis of a sea urchin gene. Science 1998; Bray et al. Receptor clustering as a cellular mechanism to control sensitivity. Nature 1998; Bhalla ad Iyengar. Emergent properties of signaling pathways. Science 1998 Structuring of the community modelling metabolism Large-scale modelling and simulation projects E-Cell project 1996; The Virtual Cell 1998 Availability of high-throughput data on parts and interactions Two-hybrids (1989); microarrays (1995) etc. Large-scale funding for wet+dry studies of biological systems Alliance For Cellular Signalling ( First of the NIH glue grants. 1998

29 Towards Systems Biology Organ level SBML XXI XX XIX 1980 XVIII Purkinje neuron Cell cycle Goldbeter/Koshland Covalent cascades Metabolic networks Complex neurons Stochastic algorithms MCA/BST Boolean networks Denis Noble Heart pacemaker Rall's cable approximation neurobiology XVI User-friendly simulators Multi-agent systems Simple circuits XVII 1960 Barabasi Tissue models Signalling pathways Gene regulatory networks Hodgkin/Huxley models Repressilator methods

30 Towards Systems Biology Synthetic Synthetic Biology Biology Organ level SBML XXI XX XIX 1980 XVIII Purkinje neuron Cell cycle Goldbeter/Koshland Covalent cascades Metabolic networks Complex neurons Stochastic algorithms MCA/BST Boolean networks Denis Noble Heart pacemaker Rall's cable approximation neurobiology XVI User-friendly simulators Network Network Multi-agent systems Simple circuits XVII 1960 Barabasi Tissue models Signalling pathways Gene regulatory networks Hodgkin/Huxley models Repressilator methods Biology Biology

31 Formal revival of Systems Biology Modelling Systems Biology Hiroaki Kitano founds the Systems Biology Institute in Tokyo First appearance: Kyoda, Kitano. Virtual Drosophila project: Simulation of drosophila leg formation. Genome Informatics Series (1998) Kitano, H. Perspectives on systems biology. New Generation Computing Volume 18, Issue 3, 2000, Pages Network Systems Biology First appearance: Leroy Hood. Systems biology: new opportunities arising from genomics, proteomics and beyond. Experimental Hematology. Volume 26, Issue 8, 1998, Page 681 Schwikowski B, Uetz P, Fields S. A network of protein-protein interactions in yeast. Nat Biotechnol Dec;18(12): Leroy Hood founds the Systems Biology Institute in Seattle

32 Rise of Systems Biology as a paradigm 2007 XXI XX XIX XVIII XVII XVI

33 Rise of Systems Biology as a paradigm 2007 XXI XX XIX ERASysBio SystemsX Systems Biology Enters FP6 hepatosys YSBN 2002 XVIII XVII Alliance for Cellular Signaling ERATO-Kitano XVI projects

34 Rise of Systems Biology as a paradigm 2007 XXI XX XIX ERASysBio SystemsX Systems Biology Enters FP6 hepatosys YSBN Choi Boogerd Grierson Palsson BMC Sys Bio Alon Szallasi Kriete MSB Klipp Kaneko IEE Sys Bio SBML Science special issue 2002 XVIII XVII Alliance for Cellular Signaling ERATO-Kitano Foundations of Systems Biology Ideker/Hood Von Dassow Computational Cell Biology ECell XVI projects publications

35 Rise of Systems Biology as a paradigm 2007 XXI XX XIX ERASysBio SystemsX Systems Biology Enters FP6 hepatosys YSBN Choi Boogerd Grierson Palsson BMC Sys Bio Alon Szallasi Kriete MSB Klipp Kaneko 6 BBSRC centres IEE Sys Bio BioQuant SBML Science special issue 2002 XVIII XVII Alliance for Cellular Signaling ERATO-Kitano Foundations of Systems Biology Ideker/Hood Tokyo Systems Biology Institute Von Dassow Computational Cell Biology Seattle Institute for Systems Biology ECell XVI projects publications Institutes

36 Rise of Systems Biology as a paradigm 2007 XXI XX XIX ERASysBio SystemsX Systems Biology Enters FP6 hepatosys YSBN Choi Boogerd Grierson Palsson BMC Sys Bio Alon Szallasi Kriete MSB Klipp Kaneko End End of of BBSRC BBSRC Systems Systems Biology Biology calls calls 6 BBSRC centres IEE Sys Bio BioQuant SBML Science special issue 2002 XVIII XVII Alliance for Cellular Signaling ERATO-Kitano Foundations of Systems Biology Ideker/Hood Tokyo Systems Biology Institute Von Dassow Computational Cell Biology Seattle Institute for Systems Biology ECell XVI projects publications Institutes

37 But what is Systems Biology??? Molecular and Cellular Biology Describes and quantify Level N A B properties

38 But what is Systems Biology??? Molecular and Cellular Biology Describes and quantify Level N A B properties A B properties C Level N+1 Describes and quantify Physiology

39 But what is Systems Biology??? Molecular and Cellular Biology Describes and quantify A Level N B properties explains Systems Biology C Level N+1 A affects B properties Describes and quantify Physiology

40 But what is Systems Biology??? Molecular and Cellular Biology Describes and quantify A Level N B properties explains Systems Biology C Level N+1 A affects B properties NOT UNIQUE! Describes and Several systems structures quantify may generate the same properties Physiology

41 What Systems Biology is not! Purely theoretical: Most systems biologists are actually experimental biologists

42 What Systems Biology is not! Purely theoretical: Most systems biologists are actually experimental biologists Based on large datasets: in a system of two enzymes, the behaviour of both reactions is different than the ones observed in isolation

43 What Systems Biology is not! Purely theoretical: Most systems biologists are actually experimental biologists Based on large datasets: in a system of two enzymes, the behaviour of both reactions is different than the ones observed in isolation Focused on biomolecular systems: systems biology is scale-free, and a biological system can be made up of molecules, cells, organs or individuals

44 What Systems Biology is not! Purely theoretical: Most systems biologists are actually experimental biologists Based on large datasets: in a system of two enzymes, the behaviour of both reactions is different than the ones observed in isolation Focused on biomolecular systems: systems biology is scale-free, and a biological system can be made up of molecules, cells, organs or individuals

45 Systems Biology is the study of the emerging properties of a biological system, taking into account all the necessary constituents, their relationships and their dynamics

46 Two kinds of Systems Biology? Systems-wide analysis (omics) Application of systems-theory Born: 1990s Born: 1960s Technologies: high-throughput, statistics Technologies: quantitative measurements, modelling People's background: molecular biologists, mathematicians People's background: biochemists, engineers Key lesson: the selection of a phenotype is done at the level of the system, not of the component (gene expression puzzle) Key lesson: the properties at a certain level are emerging from the dynamic interaction of components at a lower level

47 Procedure does not depend on directionality Bottom-up Top-down literature pathway database Put numbers biochemistry omics Parametrise parameter search hmm... Simulation structural analysis, steady-state analysis Build the system Analyse perturb Inhibition, stimulation, suppression, overexpression

48 The challenges ahead Types of representation Scales and the mesoscopic gap Genotype-system-phenotype problem Drug discovery models Vs systems modelling Drug discovery models Vs omics data

49 The four views of systems biology

50 Interaction networks Non-directional Non-sequential Non-mechanistic Statistical modelling Functional genomics IntAct, DIP, String

51 Activity-Flows RAF Directional Sequential MEK ERK Non-mechanistic Logical modelling Signalling pathways, gene regulatory networks KEGG, STKEs

52 Entity Relationships Directional Ub P K Y T X Z Non-sequential Mechanistic Independent rules: no explosion Rule-based modelling Molecular Biology MIM

53 Process Descriptions S1 Directional S2 Sequential P1 Mechanistic E Subjected to combinatorial explosion S3 P2 P Process modelling Biochemistry, Metabolic networks KEGG, Reactome

54 The four views are orthogonal projections of the underlying biological phenomena

55 A A B C B Think scaffolding proteins B A C A B C Think phosphorylated signals

56 The problem of scales QSAR chemical kinetics pharmacometrics physiology

57 The mesoscopic gap Time scale life Development Development degeneration degeneration month hour Cell Cell signalling signalling pathways pathways s ms µs behaviour behaviour tissus tissus excitability excitability reaction-diffusion reaction-diffusion Conformational ns Conformational transitions transitions ps fs Molecular Molecular dynamics dynamics nm m mm m Spatial scale

58 The mesoscopic gap Time scale life Development Development degeneration degeneration month hour Cell Cell signalling signalling pathways pathways s ms µs behaviour behaviour tissus tissus excitability excitability reaction-diffusion reaction-diffusion Conformational ns Conformational transitions transitions 24 orders of magnitude ps fs Molecular Molecular dynamics dynamics nm 9 orders of magnitude m mm m Spatial scale

59 The mesoscopic gap Time scale life Development Development degeneration degeneration month hour Cell Cell signalling signalling pathways pathways s ms 24 orders of magnitude reaction-diffusion reaction-diffusion ns Conformational Conformational transitions transitions fs tissus tissus excitability excitability µs ps behaviour behaviour Theoretical and Computational roadblockers Molecular Molecular dynamics dynamics nm 9 orders of magnitude m mm m Spatial scale

60 Multi-scale does not mean fine scale plus massive power Computational issues In 2008, largest simulations were 1 million atoms for 50 ns (using > 30 years CPU) and atoms for 0.5 ms. E-coli ~ 1 million million atoms... We need at least 1014 more power to simulate E coli during 1 s Theoretical issues Molecular dynamics does not scale linearly with number of interactions We do not know how to simulate mesoscopic phenomena: conformational transitions, secondary structure movements etc.

61 Emergent properties and the gene-system-phenotype puzzle Waddington C.H., Kacser H (1957) The Strategy of the Genes: A Discussion of Some Aspects of Theoretical Biology. George Allen & Unwin

62 Emergent properties and the gene-system-phenotype puzzle Waddington C.H., Kacser H (1957) The Strategy of the Genes: A Discussion of Some Aspects of Theoretical Biology. George Allen & Unwin physiology phenotype systems biology system genetics genotype

63 Emergent properties and the gene-system-phenotype puzzle Waddington C.H., Kacser H (1957) The Strategy of the Genes: A Discussion of Some Aspects of Theoretical Biology. George Allen & Unwin Many networks can theoretically generate the same phenotype, and this happens, in a synchronous (sister cells with same phenotype but different transcript/prote/metabol/omes) and diachronous manner (*omes of a cell changes over time but same phenotype).

64 Reverse engineering is hard... Gene A Gene B Gene C Gene... D... Phenotype X Phenotype Y Phenotype Z... Gene A? Gene B... Gene C... Gene D

65 Drug discovery modelling: pharmacometrics in(t) C k dc = in(t) k C dt C pat1 pat2 pat time C Cmax tmax time

66 Drug discovery modelling: pharmacometrics PK PD effect [druga] time [druga]

67 Drug discovery modelling: pharmacometrics PK PD effect [druga] [druga] time effect [drugb] time [drugb]

68 Drug discovery modelling: pharmacometrics PK PD effect [druga] [druga] time Effect of A+B? effect [drugb] time [drugb]

69 Systems modelling D1 T d[t ] = k1 [T ] [D1] dt D1 [T] T time d[t ] = k2 [T ] [D2] dt T D2 [T] D2 T time

70 Systems modelling d[t ] = k1 [T ] [D1] k2 [T ] [D2] dt D1 D1 T [T] T D2 time D2 T

71 Drug discovery and pharmacometrics models in(t) C k

72 Drug discovery and pharmacometrics models in(t) We (think we) know how to do that C k

73 Drug discovery and omics?

74

75