the erato systems biology workbench

25
The ERATO Systems Biology Workbench Michael Hucka, Hamid Bolouri, Andrew Finney, Herbert Sauro ERATO Kitano Systems Biology Project California Institute of Technology, Pasadena, CA, USA Principal Investigators: John Doyle, Hiroaki Kitano Collaborators: Adam Arkin (BioSpice), Dennis Bray (StochSim), Igor Goryanin (DBsolve), Les Loew (Virtual Cell), Pedro Mendes (Gepasi/Copasi), Masaru Tomita (E-Cell)

Upload: osgood

Post on 31-Jan-2016

35 views

Category:

Documents


0 download

DESCRIPTION

The ERATO Systems Biology Workbench. Michael Hucka, Hamid Bolouri, Andrew Finney, Herbert Sauro ERATO Kitano Systems Biology Project California Institute of Technology, Pasadena, CA, USA Principal Investigators : John Doyle, Hiroaki Kitano - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: The ERATO Systems Biology Workbench

The ERATO Systems Biology Workbench

Michael Hucka, Hamid Bolouri,Andrew Finney, Herbert Sauro

ERATO Kitano Systems Biology ProjectCalifornia Institute of Technology, Pasadena, CA, USA

Principal Investigators: John Doyle, Hiroaki Kitano

Collaborators: Adam Arkin (BioSpice), Dennis Bray (StochSim), Igor Goryanin (DBsolve), Les Loew (Virtual Cell), Pedro Mendes

(Gepasi/Copasi), Masaru Tomita (E-Cell)

Page 2: The ERATO Systems Biology Workbench

2

Background• Modeling, simulation & analysis are critical

– Huge volumes of data– Many disparate findings

• Rapid rate of software tool development– Roles: data filtering, model creation, model simulation – Many groups are creating many tools

• Different packages have different niche strengths reflecting expertise & preferences of the group

• Strengths are often complementary to those of other packages

Page 3: The ERATO Systems Biology Workbench

3

Complementary Strengths of Tools

Tool 7

Tool 3

Tool 4

Tool 2

Tool 6

Tool 5

Tool 1

Multistate reactions/stochastic

Reaction/Diffusion

Visualization of networks

Optimization

Bifurcation analysis

Handle large systems

Page 4: The ERATO Systems Biology Workbench

4

Problems• No single package answers all needs of modelers

• No single tool is likely to do so in the near future– Range of capabilities is large– New techniques ( new tools) evolving too rapidly

• Researchers are likely to continue using multiple packages for the foreseeable future

• Problems in using multiple tools:– Simulations & results often cannot be shared or re-used – Duplication of software development effort

Page 5: The ERATO Systems Biology Workbench

5

Goal & Approach• Systems Biology Workbench goal: to provide

software infrastructure that – Enables sharing of simulation/analysis software & models

– Enables collaboration between software developers

• Two-pronged approach:– Develop a common model exchange language

• SBML: Systems Biology Markup Language– Develop an environment that enables tools to interact

• SBW: Systems Biology Workbench

Page 6: The ERATO Systems Biology Workbench

6

Systems Biology Markup Language

• Problem:– Many software tools, few common exchange formats

• Difficult to take advantage of multiple tools• Difficult to establish repositories of models

• A Solution (In Principle):– Define a common exchange language

• Use a simple, well-supported, textual substrate (XML)• Add components that reflect the natural conceptual

constructs used by modelers in the domain

Page 7: The ERATO Systems Biology Workbench

7

• Domain: biochemical network models

• A model is described using a list of components:

– Beginning of model definition» List of unit definitions (optional)» List of compartments» List of species» List of parameters (optional)» List of rules (optional)» List of reactions

– End of model definition

• Each component has a specific structure

Structure of Model Definitions in SBML

Page 8: The ERATO Systems Biology Workbench

8

Example

S1

X2

X1

k1*X0

k2*S1

k3*S1

X0

Page 9: The ERATO Systems Biology Workbench

9

Example (cont.)<?xml version=“1.0” encoding=“UTF-8”?><sbml level=“1” version=“1”> <model name=“simple”>

<listOfCompartments> <compartment name=“c1” /> </listOfCompartments>

<listOfSpecies> <specie name=“X0” compartment=“c1” boundaryCondition=“true” initialAmount=“1”/> <specie name=“S1” compartment=“c1”

boundaryCondition=“false” initialAmount=“0”/> <specie name=“X1” compartment=“c1”

boundaryCondition=“true” initialAmount=“0”/> <specie name=“X2” compartment=“c1”

boundaryCondition=“true” initialAmount=“0.23”/> </listOfSpecies>

Page 10: The ERATO Systems Biology Workbench

10

Example (cont.)<?xml version=“1.0” encoding=“UTF-8”?><sbml level=“1” version=“1”> <model name=“simple”>

<listOfCompartments> <compartment name=“c1” /> </listOfCompartments>

<listOfSpecies> <specie name=“X0” compartment=“c1” boundaryCondition=“true” initialAmount=“1”/> <specie name=“S1” compartment=“c1”

boundaryCondition=“false” initialAmount=“0”/> <specie name=“X1” compartment=“c1”

boundaryCondition=“true” initialAmount=“0”/> <specie name=“X2” compartment=“c1”

boundaryCondition=“true” initialAmount=“0.23”/> </listOfSpecies>

Page 11: The ERATO Systems Biology Workbench

11

Example (cont.)<?xml version=“1.0” encoding=“UTF-8”?><sbml level=“1” version=“1”> <model name=“simple”>

<listOfCompartments> <compartment name=“c1” /> </listOfCompartments>

<listOfSpecies> <specie name=“X0” compartment=“c1” boundaryCondition=“true” initialAmount=“1”/> <specie name=“S1” compartment=“c1”

boundaryCondition=“false” initialAmount=“0”/> <specie name=“X1” compartment=“c1”

boundaryCondition=“true” initialAmount=“0”/> <specie name=“X2” compartment=“c1”

boundaryCondition=“true” initialAmount=“0.23”/> </listOfSpecies>

Page 12: The ERATO Systems Biology Workbench

12

Example (cont.)<listOfReactions>

<reaction name=“reaction_1” reversible=“false”> <listOfReactants> <specieReference specie=“X0” stoichiometry=“1”/> </listOfReactants> <listOfProducts> <specieReference specie=“X0” stoichiometry=“1”/> </listOfProducs> <kineticLaw formula=“k1 * X0”> <listOfParameters> <parameter name=“k1” value=“0”/> </listOfParameters> </kineticLaw> </reaction>

<reaction name=“reaction_2” reversible=“false”> <listOfReactants> <specieReference specie=“S1” stoichiometry=“1”/> </listOfReactants> . . .

Page 13: The ERATO Systems Biology Workbench

13

Some Points About SBML• Users do not write in XML — software tools do!• SBML is being defined incrementally

– SBML Level 1 covers non-spatial biochemical models• Intentionally kept simple for maximal compatibility

– SBML Level 2 will extend Level 1 with more facilities

• Defined in abstract form (UML) + textual descriptions– Used to define XML encoding + XML Schema

Level 2

Level 3?

Level 1

E.g.: • Composition• Geometry• Arrays… others

Page 14: The ERATO Systems Biology Workbench

14

Systems Biology Workbench• Open-source, integrated software environment

that enables sharing of computational resources– Allows software developers to build interprocess

communications facilities into their applications

• From the user’s perspective:– One SBW-enabled application can interact with another– Each application or module offers services to others

• E.g.: ODE solution, time-based simulation, visualization, etc.

Page 15: The ERATO Systems Biology Workbench

15

From the User’s Perspective

Page 16: The ERATO Systems Biology Workbench

16

From the User’s Perspective

Page 17: The ERATO Systems Biology Workbench

17

From the User’s Perspective

Page 18: The ERATO Systems Biology Workbench

18

Behind the Scenes

Page 19: The ERATO Systems Biology Workbench

19

From the Programmer’s Perspective

• Desirable features– Small application programming interface (API)– Libraries that implement inter-program communications– A registry of services for applications to query– XML-based model representation (SBML)

• Uses well-known, proven technologies– Communications via message-passing over plain sockets– Modular, distributed, broker-based architecture

• API provides two styles:– “Low-level”: call/send operations + directory services– “High-level”: object-oriented interface layered on top

Page 20: The ERATO Systems Biology Workbench

20

Driving Principles• Keep it language-neutral

– We’ll provide C++, Java, Delphi libraries for Win, Linux• But can be implemented in any language

• Keep it simple– Simple message-passing scheme

• Avoid complexity & size of CORBA• Easy to make cross-platform compatible• Easy to make distributed

– Simple low-level API, convenient higher-level API

• Make sure contributors benefit– Open source development– Symmetric infrastructure: no application dominates

Page 21: The ERATO Systems Biology Workbench

21

Modules Planned• Data filtering & preparation

• Database support– E.g.: web searching, storage management

• Model definition & manipulation– E.g.: scripting languages, visual editors

• Equation solvers– E.g.: ODEs, DAEs, stochastic

• Analysis & visualization tools– E.g.: bifurcation, 2-D/3-D/4-D plotting

• Optimization & parameter searching

Page 22: The ERATO Systems Biology Workbench

22

Summary & Availability• SBML

– Level 1 specification is publicly available• http://www.cds.caltech.edu/erato

– Support being added by other groups to their apps

• SBW– Preliminary test implementation completed– Production version is now in development

• Draft API definition & other info available– http://www.cds.caltech.edu/erato/sbw/docs

– Expect first public release in November atICSB 2001 (http://www.icsb2001.org)

Page 23: The ERATO Systems Biology Workbench

23

Page 24: The ERATO Systems Biology Workbench

24

Outline1) Background and motivations

I.e., why are we doing this?

2) Systems Biology Markup Language (SBML)

Model description language & exchange format

3) Systems Biology Workbench

Infrastructure for connecting software tools

Page 25: The ERATO Systems Biology Workbench

25

Scope of SBML• Domain: biochemical network models

S1 S2 S3X0 X1

K1*X0 - S1*K2

S3/Ki

Vmax*S1

Km+S1

k3*S2

k4*S2 - k5*S3 k6*S3

k7*S3*S12

X2