Peyer's Patch Development - Computer Simulation

Included within the submission:
"Pairing experimentation and computational modelling to understand the role of tissue inducer cells in the development of lymphoid organs"

Kieran Alden, Jon Timmis, Paul S Andrews, Henrique Veiga-Fernandes, Mark C Coles

Skip to: Domain Model Platform Model Assumptions Simulation



A. Domain Model The domain model captures an abstraction of the biological domain. The state diagrams, created using a modified version of Unified Modelling language, provide a description of the states in which the identified agents (cell type) may exist within (the boxes), and the biological event that must take place for that agent to transition into the next state (the arrow). This does not contain any simulation-specific detail.

An Activity Diagram also forms part of this model, representing the low-level interactions between the cells (LTin, LTi, LTo Cell) which lead to the formation of PP. Cellular behaviours are described in boxes, decisions points indicated by diamonds and lines. Arrows indicate potential changes to cellular behaviour.

Once loaded, click on the diagram to enlarge
View Domain Model State Diagrams
View Activity Diagram
B. Platform Model The Platform model details how the states and interactions captured in the domain model are coded into the simulation. The expected behaviour, which emerges from interactions between components in the system, is specifically not present in the platform model. Behaviour emerges from the simulation and are not coded into the simulation. In the platform model, how each cell behaves and how interactions are encoded is detailed. As this includes a variety of factors from the domain model a number of assumptions are made and documented.

Once loaded, click on the diagram to enlarge
View Platform Model State Diagrams
Intestine Environment Representation
Adhesion Factors in the Platform Model
Chemokines in the Platform Model
Table of additional considerations included in the Platform Model
Table of Parameters Identified in Domain and Platform Model
C. Assumptions Made Although in theory including all current understanding of a system is possible, such a model would be very difficult to construct and is unlikely to significantly improve the model. Therefore, the domain model is an abstraction of the biological system, with suitable assumptions made where necessary and clearly documented for later scrutiny.

Further parameters are identified in forming the Platform Model and described, and as in the domain model the numerical values of some parameters are unknown. These parameters may capture the behaviour of component parts such as adhesion molecules, cytokines, and chemokines. Despite the importance of these factors, the number of molecules expressed by the different cell types, the level of chemokine expression required to induce cellular chemotaxis and the diffusion distribution of chemokines and cytokines in the localised environment all are currently unknown. Thus further assumptions are made based on known biology and documented.

View Assumptions Tables for this Model
D. The Simulator The simulation can be run either with the default parameter values (set during calibration, leading to cell behaviour that is statistically similar to that seen ex vivo), or these can be set prior to the simulation run.


Tool Information

The following applet loads the entire MASON toolkit so it may take a second. Please note:

  • There are known problems with some Mac web browsers, as their Java may not be hardware-accelerated and the 2D graphics is very slow. You'll have more luck running MASON directly as a Java 1.3.1 application. See the main web page for more information.
  • Mac web browsers except for Safari operate best when the Java 1.4.2 plugin is running.
  • Once a simulation is running, you can start another simulation under the File menu.
  • If you close a display, it is hidden -- you can get it back under the Displays tab in the console window. If you close the the console window, the simulation is quit.