Knowledge of a given pathway system is usually available in descriptive form as text or as diagrammatic representations in papers or reviews. Increasingly, information and pathway diagrams can also be found in interaction and pathway databases. In order to create a graphical representation of a pathway and thereby the beginnings of computational model, one needs to extract the relevant information from the literature or pathway databases.
STEP 1: Gather information about the pathway that you want to model: read papers, look at existing pathway diagrams.
STEP 2: Identify the biological components of the pathway (e.g. proteins, nucleic acids, etc.) and the interactions between them. Identify the cellular compartments and cell types in which the interactions take place (e.g. nucleus, mitochondrion, etc.). Record the source of the information.
Pathway Description (Components – Interactions – Compartments)
Interferon beta (IFNB1) homodimerises and binds to its cell surface receptor complex, composed of the transmembrane proteins IFNAR1 and IFNAR2 and the intracellular kinases TYK2 and JAK1. The complex is composed of 2 of each of these proteins. Binding causes a conformational change in the complex, resulting in the autophosphorylation of JAK1. Once activated, the complex catalyses the phosphorylation of STAT2, which forms a heterodimer with STAT1. This complex then binds interferon regulatory factor 9 (IRF9) forming the complex often referred to as ISGF3 and translocates to the nucleus. Here it activates the transcription of a number of genes including IRF2, IL12B, STAT1, IL15, TAP1, GBP1, PSMB9, and SOCS3. In turn, SOCS3 inhibits the autophosphorylation of the receptor, thereby preventing further activation.
Pathways can be drawn and edited using the yEd graphic editor following the modified Edinburgh Pathway Notation (mEPN) scheme (Freeman et al. 2010) a graphical notation system based loosely on the concepts of the process diagram (Kitano et al. 2005). STEP 1: Download and install yEd
- Go to the yEd website and download the software (it’s free!)
- Download the mEPN palette file: mEPN notation scheme 2014 Palette
- Load the mEPN palette file into the yEd palette. Menu: Edit → Manage Palette… → Import Selection
STEP 2: Familiarise yourself with the yEd software Play around with the software; it’s relatively intuitive and easy to use. STEP 3: Create pathway components nodes Components of pathways are represented using Entity Nodes (see mEPN notation). The appropriate component node should be selected from the palette and the name of the component should be added. A hyperlink can be added pressing [F6] to access the node’s properties dialogue. As a quick start, you can download a GraphML file of ready-made pathway components and use these: Interferon Beta Pathway Components. Alternatively, click the button to open yEd with the file preloaded:
STEP 4: Draw the interactions between components The molecular interactions between pathway components can be represented using Process Nodes to describe processes (e.g. binding) and Edges that join Entity and Process nodes denoting the direction andtype of interaction (e.g. inhibition) (see mEPN Notation (link)). Add compartment nodes. Ensure the bipartite graph structure is maintained (component-process-component). The figure below shows the Interferon Beta pathway drawn with the mEPN Notation. The model contains a feedback loop. Download a ready-made GraphML file for this model: Interferon Beta Pathway. Alternatively, click the button to open yEd with the file preloaded:
Users provide information about starting conditions of the pathway through the placement of tokens that indicate the quantity of each component. The parameterisation of the model ideally requires some experimental data providing information on the relative initial concentrations of the pathway components where known. The dynamic behaviour of the pathway is then simulated by the movement of these tokens between Entity nodes (token flow). STEP 1: Place tokens to define the initial state Place input tokens on the Edge prior to entity nodes to indicate the availability of that component within the cell at the start of the simulation. Token values represent a components relative concentration or activity. STEP 2: Save pathway in GraphML file format The figure below shows a Signalling Petri Net model of the Interferon Beta pathway drawn with the mEPN Notation, containing a feedback loop and parameterised with tokens.
Download a ready-made GraphML file for this model: IFNB Pathway SPN Feedback. Alternatively, click the button to open yEd with the file preloaded:
mEPN pathway diagrams, saved in yEd in the GraphML file format can be loaded into Graphia Professional in order to visualise the pathway in a 3D environment and perform stochastic flow simulations using a modified version of Signalling Petri Net (SPN) algorithm.
STEP 1: Install Graphia Professional
Go to Kajeka website download program and install.
STEP 2: Set-up the simulation
Open a GraphML file of a mEPN diagram in Graphia Professional. The SPN dialogue is displayed, in which the user can select options for the simulation algorithm.
The user can decide how many times to tokens are moved through the system (Number of Time Blocks, in a time block all transitions are fired once at random).
The simulation can be repeated multiple times (Number of Runs) and runs are averaged to calculate variance.
The user can also select the type of probabilistic distribution of tokens flow. Each time a transition is fired, the number of tokens moved between input and end nodes can be completely random (Uniform Distribution) or chosen from a Standard Normal distribution (Average 50% of the input tokens is moved).
STEP 3: Save and view the simulation results
Once the SPN simulation algorithm has finished, a Simulation Results dialogue appears allowing users to save the simulation results as .spn or .txt file. You can place the mouse over a node to visualise the flow of tokens through that node.
Users can control the animation of the simulation with the Simulation Animation Control dialogue to change the animation speed, nodes size and colour.
Token flow through the mEPN diagram may then be animated, with inflation and contraction of the component nodes representing the accumulation and degradation of reactants in the pathway. In this way, complex biological processes with many interacting reactants may be modelled.