Progress WP4, WP7 Memodyn Application Scenario Organizational - - PowerPoint PPT Presentation

▶
progress wp4 wp7
SMART_READER_LITE
LIVE PREVIEW

Progress WP4, WP7 Memodyn Application Scenario Organizational - - PowerPoint PPT Presentation

Oct 25, 2022 283 likes •734 views

Progress WP4, WP7 Memodyn Application Scenario Organizational Coarse-Graining Jan Huwald Richard Henze Bashar Ibrahim Peter Dittrich Bio Systems Analysis Group, Institute of Computer Science, Friedrich-Schiller-University Jena

slide-1
SLIDE 1

Progress WP4, WP7

Memodyn – Application Scenario – Organizational Coarse-Graining

Jan Huwald Richard Henze Bashar Ibrahim Peter Dittrich

Bio Systems Analysis Group, Institute of Computer Science, Friedrich-Schiller-University Jena

Application scenario in collaboration with: Diekmann Group and Hemmerich Group, FLI Jena

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 1

slide-2
SLIDE 2

Overview

I. Memodyn (WP4)

Report in D4.2 Software / source code in D4.2

  • II. Artificial chemistries and organizational coarse-

graining (WP4, WP7)

Paper in D4.2: Keyssig et al., Bioinformatics, 2014

  • III. Application scenario – Mitotic checkpoint (WP7)

 Paper in D7.1: Ibrahim&Henze, Int. J. Mol. Sci., 2014  Paper in D7.1: Henze et al., Biosystems, 2014

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 2

slide-3
SLIDE 3

Part I MEMODYN

WP 4 Jan Huwald

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 3

slide-4
SLIDE 4

Memodyn

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 4

slide-5
SLIDE 5

Memodyn „Learning Cycle“

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 5

slide-6
SLIDE 6

Continuous sample generation

  • 1. Random unbiased sampling
  • 2. Constraint propagation
  • 3. Force-based search

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 6

slide-7
SLIDE 7

Unbiased random sampling

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 7

slide-8
SLIDE 8

Constraint propagation

(using GECODE)

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 8

slide-9
SLIDE 9

Force-based approach

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 9

slide-10
SLIDE 10

Force-based approach

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 10

slide-11
SLIDE 11

Force-based approach

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 11

slide-12
SLIDE 12

Force-based approach

For all combination all solutions are found!

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 12

slide-13
SLIDE 13

Continuous simulation

  • 1. Micro-simulation border
  • 2. Energy guided distance

quantization

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 13

slide-14
SLIDE 14

Micro-simulation border

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 14

slide-15
SLIDE 15

Micro-simulation ersatz field

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 15

slide-16
SLIDE 16

Micro-simulation border for three particles

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 16

slide-17
SLIDE 17

Software

Source code implementing the methods mentioned above attached to D4.2.

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 17

slide-18
SLIDE 18

Part II Organizational Coarse-Graining

WP 4 / WP 7

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 18

slide-19
SLIDE 19

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 19

  • 1. Basic Idea

1 3 2 4

Chemical Organization Theory Organizations Reaction network

1 3 2 4

Organization

[P. Dittrich, P. Speroni di Fenizi, Chemical Organization Theory, Bull. Math. Biol., 2007]

slide-20
SLIDE 20

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 20

  • 1. Basic Idea

{1} {2, 3} {1,2,3,4} { }

Hasse diagram of

  • rganizations

Organizations Reaction network Chemical Organization Theory

1 3 2 4

[P. Dittrich, P. Speroni di Fenizi, Chemical Organization Theory, Bull. Math. Biol., 2007]

slide-21
SLIDE 21

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 21

  • 1. Basic Idea

1 3 2 4

Thoerie chemischer Organization

{1} {2, 3} {1,2,3,4} { }

Dynamics

[2] [3] [4] [1]

Chemical Organization Theory

Hasse diagram of

  • rganizations

Organizations Reaction network

[P. Dittrich, P. Speroni di Fenizi, Chemical Organization Theory, Bull. Math. Biol., 2007]

slide-22
SLIDE 22

Organisational Coarse-graining Results

  • 1. Discrete organizations (Kreyssig et al. 2014)
  • 2. Measuring organizations
  • 3. Hierarchical dynamics

(Boolean / Neuronal networks example)

slide-23
SLIDE 23
  • 1. Discrete organization

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 23

[Kreyssig et al., Bioinformatics, 2014]

slide-24
SLIDE 24
  • 1. Discrete Organizations

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 24

[Kreyssig et al., Bioinformatics, 2014]

2 C 

slide-25
SLIDE 25
  • 1. Discrete Organizations

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 25

[Kreyssig et al., Bioinformatics, 2014]

{A, B, C} is a purely discrete organization

2 C 

slide-26
SLIDE 26
  • 2. Measuring Organizations

So far, a reaction network was necessary. Now, measuring organizations and the hierarchical organizational structure directly, without the need to identify individual species or reactions.  “natural” coarse-graining, since it is derived from (physical) measurements

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 26

slide-27
SLIDE 27
  • 2. Measuring Organizations - Recipe

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 27

slide-28
SLIDE 28
  • 2. Measuring Organizations - Status
  • Basic theory ready
  • Prototypic software ready
  • 16 species artificial chemistry with 50
  • rganizations ready for testing.

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 28

slide-29
SLIDE 29
  • 3. Hierarchical dynamics Boolean and

neural networks

Is there a hierarchy of attractors? How is the hierarchy of attractors related to the hierarchy of organizations?  Attractors and organizations for coarse-graining

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 29

[SORN Network by Triesch et al. ]

slide-30
SLIDE 30
  • 3. Hierarchy of attractors

[Lukas Klimmasch] - Example

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 30

Subset of active neurons Set of active neurons within one attractor = “brain region”

slide-31
SLIDE 31
  • 3. Another Hierarchy of Attractors

[Lukas Klimmasch] - Example

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 31

slide-32
SLIDE 32

Part III Application Scenario

WP 7 Richard Henze, Bashar Ibrahim

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 32

slide-33
SLIDE 33

16.11.06, Asselsheim Peter Dittrich (FSU Jena) 33

Wait until all kinetochores are correctly attached

http://library.thinkquest.org/C004535/mitosis .html

WAIT!

slide-34
SLIDE 34

16.11.06, Asselsheim Peter Dittrich (FSU Jena) 34

Wait until all kinetochores are correctly attached

http://library.thinkquest.org/C004535/mitosis .html

GO!

slide-35
SLIDE 35

16.11.06, Asselsheim Peter Dittrich (FSU Jena) 35

Wait until all kinetochores are correctly attached

http://library.thinkquest.org/C004535/mitosis .html

GO!

slide-36
SLIDE 36

Application Scenario Results

  • 1. Preliminary studies:

a) Active transport of Mad2 (Ibrahim&Henze, Int. J. Mol. Sci., 2014) b) Rule-based modeling of kinetochore mutants. (Henze et al, Biosystems, 2014)

  • 2. Checkpoint Scenario
  • 3. PML nuclear bodies (potentially another

application scenario)

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 36

slide-37
SLIDE 37
  • 1. Preliminary Work and Results

Various models at different scales of coarse- graining now available. From ODE to detailed rule-based spatial particle simulation.

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 37

slide-38
SLIDE 38
  • 1. Example: 3-D Rule-Based Model of Full Kinetochore

38 Henze / Dittrich et al. - FSU Jena 11.12.2014 Brussels, HIERATIC

by R. Henze, B. Ibrahim, et al., FSU Jena, 2014

slide-39
SLIDE 39
  • 2. Mitotic Checkpoint Scenario

Microlevel: Simulate all kinetochores (92), in a realistic 3D space, and realistic particle numbers (1 Mio) of inhibitors and activators. Why: To have a trustworthy model that is

  • Understandable by domain experts (biologists)
  • Takes directly the rules from domain experts

Drawback: Computation time  Coarse graining need for efficient computation, e.g., if predictions.  Also getting additional understanding, by extracting more general laws.

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 39

slide-40
SLIDE 40
  • 2. Mitotic Checkpoint Scenario

Unatched-Kinetochor  activates Inhibitor No inhibitor around a kinetochore  GO signal

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 40

slide-41
SLIDE 41

Start

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 41

slide-42
SLIDE 42

End

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 42

slide-43
SLIDE 43

Preliminary simulation of mitotic checkpoint

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 43

slide-44
SLIDE 44

Summary

I. Memodyn (WP4)

Report in D4 Software / source code in D4.2

  • II. Artificial chemistries and organizational coarse-

graining (WP4, WP7)

Paper in D4.2: Keyssig et al., Bioinformatics, 2014

  • III. Application scenario – Mitotic checkpoint (WP7)

 Paper in D7.1: Ibrahim&Henze, Int. J. Mol. Sci., 2014  Paper in D7.1: Henze et al., Biosystems, 2014

11.12.2014 Brussels, HIERATIC Henze / Dittrich et al. - FSU Jena 44