CellML, PMR, OpenCOR, CRBM, David Nickerson Auckland - - PowerPoint PPT Presentation

▶
cellml pmr opencor crbm
SMART_READER_LITE
LIVE PREVIEW

CellML, PMR, OpenCOR, CRBM, David Nickerson Auckland - - PowerPoint PPT Presentation

Jul 23, 2023 456 likes •849 views

https://doi.org/10.17608/k6.auckland.10080263 CellML, PMR, OpenCOR, CRBM, David Nickerson Auckland Bioengineering Institute University of Auckland New Zealand CellML signalling modules for the cardiac myocyte ms seconds hours days

slide-1
SLIDE 1

CellML, PMR, OpenCOR, CRBM, …

David Nickerson Auckland Bioengineering Institute University of Auckland New Zealand

https://doi.org/10.17608/k6.auckland.10080263

slide-2
SLIDE 2

Cardiac myocyte

α-adrenergic Muscarincic, ACh PLCβ Gα Gβγ PIP2 Gq α1 DAG + IP3 Ca2+ CaN CaMK PKCα PKCδ PKCβ PKCε β-adrenergic NE, Iso PKA ATP cAMP Gα Gα Gβγ AC Gi β2 Gs β1 MEF2C GATA4 NFAT p p p TFs t V Ca2+ EC-coupl.g & mechs PLB Serca2 TnC Ca2+ RyR2 TnI MHC p p p p T p p p p p ms seconds hours days

Time scale Nucleus Membrane Cytosol DNA

Inputs Outputs

c-myc, c-fos, c-jun, ras, hsp-70 TFs p Eccentric hypertrophy Concentric hypertrophy Physiological hypertrophy

CellML signalling modules for the cardiac myocyte

Ion channels, transporters INa Na+ INa,b Na+ ICl Cl- ICa,L Ca2+ ICa,T Ca2+ ICa,b Ca2+ IKr K+ IKs K+ IKto K+ IKp K+ IK1 K+ NCX Ca2+ 3Na+ 3K+ 2Na+ NKA CHE OH- NHE H+ NBC HCO3

  • Na+

AE HCO3

  • Cl-

* NO sGC sGC eNOS nNOS Ca2+ NO cGK I p cGK II p cGMP pGC ANP BNP iNOS NO Peptide GFs RAS MAPK ERK JNK p38K TFs RTK p p p p p p p Apoptosis Cytokines cytokine receptor gp130 STAT Jac IκB NFκB Inactive Class II HDACs Ca2+ CaMK PKD PKC histone Insulin, IGF, GH TFs mTOR p GSK3 p PKB PI3K PIP3 RTK p Hypertrophic cardiac myopathy FOXO

slide-3
SLIDE 3

https://cellml.org/

slide-4
SLIDE 4

4

  • XML format for encoding mathematical models
  • Reproducibility

– Unambiguous description of the mathematical model

  • Reusability

– Modular, composable

  • Comprehensible

– Metadata to describe the biological semantics

  • Tool support

– CellML API library and service – Most tools don’t support model composition

https://doi.org/10.17608/k6.auckland.10080263

slide-5
SLIDE 5

2 .0

  • Reactions are gone!
  • Only CellML allowed in the XML document

– No metadata, annotations, cmeta: id – No extension elements

  • XML syntax simplifications

– Grouping replaced with only encapsulation – No more map_components

  • Improved reusability

– Connections no longer have direction – Single interface attribute controlling scope: public, private, public_and_private, none

5

https://doi.org/10.17608/k6.auckland.10080263

slide-6
SLIDE 6
  • Units clarifications

– No need to specify base_units explicitly – Units with offsets removed – “celsius” removed from built-in units – Component-scope unit definitions removed

  • Reset rules

– Arbitrary rules to “reset” variables

  • New and compulsory MathML subset

– No more “recommended” subset to support – Well defined, no confusion

6

2 .0

https://doi.org/10.17608/k6.auckland.10080263

slide-7
SLIDE 7

lib

  • New C+ + library to meet the needs of users
  • Supporting CellML 2.0 and beyond
  • Much more streamlined and maintainable
  • Better suited for testing out new features and extensions to the

specification – Allowing rapid prototyping – Exploring alternatives – Testing model exchange and reproducibility

7

https://doi.org/10.17608/k6.auckland.10080263

slide-8
SLIDE 8

The Physiom e Model Repository – PMR

https:/ / m odels.physiom eproject.org/

  • Over 800 publicly available workspaces

– Version control repositories (git) – Historically mostly CellML models from the literature – Gradually getting more non-CellML data contributed (SED-ML, FE models, code)

  • Many more exposures

– “releases” of workspaces – A specific version processed for display and interaction

8

https://doi.org/10.17608/k6.auckland.10080263

slide-9
SLIDE 9

9

https: / / models.physiomeproject.org/ e/ 71

https://doi.org/10.17608/k6.auckland.10080263

slide-10
SLIDE 10

10

https://doi.org/10.17608/k6.auckland.10080263

slide-11
SLIDE 11

The Physiom e Model Repository – PMR

  • Consistent browser and tool integration

– Content type negotiation – Same URL – REST

  • RDF triplestore

– Indexing versioned annotations – Supporting (semantic) querying

  • Tools for model composition, parameter estimation, etc.

11

https://doi.org/10.17608/k6.auckland.10080263

slide-12
SLIDE 12

A m odelling environm ent for reproducible science https:/ / opencor.w s

12

https://doi.org/10.17608/k6.auckland.10080263

slide-13
SLIDE 13

13

https://doi.org/10.17608/k6.auckland.10080263

slide-14
SLIDE 14

14

https://doi.org/10.17608/k6.auckland.10080263

slide-15
SLIDE 15

15

https://doi.org/10.17608/k6.auckland.10080263

slide-16
SLIDE 16

16

https://doi.org/10.17608/k6.auckland.10080263

slide-17
SLIDE 17

Hands on tutorial

  • Using OpenCOR to explore modularity and reuse with CellML models

(including SED-ML)

  • Making use of PMR as a version controlled workspace to archive and share

your work FAIRly

  • Python-enabled OpenCOR
  • Starting to explore what is possible with machine learning using TensorFlow,

CellML, OpenCOR, and Python.

17

https://doi.org/10.17608/k6.auckland.10080263

Alan Garny Gonzalo Maso Talou

slide-18
SLIDE 18

18

https://reproduciblebiomodels.org/

https://doi.org/10.17608/k6.auckland.10080263

slide-19
SLIDE 19

19

Center Team

Jonathan Karr Mount Sinai TR&D 1 John Gennari U Washington TR&D 2 Ion Moraru UConn Health TR&D 3 Herbert Sauro U Washington Director David Nickerson ABI Curation Service

Support by NIBIB and NIGMS:

slide-20
SLIDE 20

20

Goals

Long-term

  • Enable more comprehensive and more predictive models that advance

precision medicine and synthetic biology Short-term

  • Make modeling more reproducible, comprehensible, reusable,

composable, collaborative, and scalable

  • Develop technological solutions to the barriers to modeling
  • Integrate the technology into user-friendly solutions
  • Push researchers to use these tools
  • Partner with journals
slide-21
SLIDE 21

21

TR&Ds

slide-22
SLIDE 22

22

Training and dissemination

slide-23
SLIDE 23

23

Curation service

slide-24
SLIDE 24

Acknow ledgem ents

  • Gonzalo Maso Talou
  • Tommy Yu
  • Alan Garny
  • Peter Hunter
  • ABI Physiome Group

Aotearoa Foundation

https://doi.org/10.17608/k6.auckland.10080263

slide-25
SLIDE 25
slide-26
SLIDE 26

26

ModelXchange

slide-27
SLIDE 27

27

SED-ML Motivation

Simulation tool

models

Biological publication repository

?

Simulation result

slide-28
SLIDE 28

Exam ple

28

First attempt to run the model, measuring the spiking rate v over time

 load SBML into the simulation tool COPASI  use parametrisation as given in the SBML file  define output variables (v)  run the time course

1 ms (standard) 100ms 1000ms

slide-29
SLIDE 29

29

Fig.: COPASI simulation, duration: 140ms, step size: 0.14

Second attempt to run the model, adjusting simulation step size and duration

Exam ple

slide-30
SLIDE 30

30

Fig.: COPASI, adjusted parameter values (a=0.02, b=0.2 c=-55, d=4)

Third attempt to run the model, updating initial model parameters

Exam ple

slide-31
SLIDE 31

31

https://sed-ml.org/

slide-32
SLIDE 32

http://co.mbine.org

slide-33
SLIDE 33

Core Standards

Standards for Visual Representation Standards for Models and their Analyses Standards for Knowledge Representation

Associated Standards

Used by core standards

Controlled Vocabularies Infrastructure Projects

BioModels.net qualifiers

BioPAX

slide-34
SLIDE 34
  • Coordination board
  • Coordinating new efforts,

meetings, etc.

○

COMBINE Archive

○

Harmonizing annotation

○

Uncertainty?

  • Publications
  • Forums/mailing lists
  • FAIR and FAIRsharing
slide-35
SLIDE 35

https://doi.org/10.1109/WSC.2017.8247840

slide-36
SLIDE 36

http://co.mbine.org/comm

slide-37
SLIDE 37
  • 10th COMBINE Anniversary
  • July 15-19 in Heidelberg
  • Registration now open!
  • Abstract submission deadline extended to June 15!
  • http://co.mbine.org/events/COMBINE_2019