The simplest simulation of the behaviour of GRNs possible in NetBuilder is as Boolean networks. Although Boolean simulations have very low resolution, and can introduce 'artefacts' such as oscillations, they are easily performed and understood. Furthermore, a Boolean simulation illustrates the basic concepts that underlie more complex simulation frameworks.
In a Boolean network, each node is in one of two states, 'on' or 'off' (also called 'true' or false', 'high' or low', and here: 'active' or 'inactive'). To indicate its state, each node has an associated value, 1 for 'on', and 0 for 'off'. During a simulation, NetBuilder goes through the topologically sorted list of nodes, looks at the input for each node and calculates its output on the basis of its input. Thus, if one of the inputs to an 'Or' function is 1, the output - or associated value - of the node is set to 1. This value is now used as input to the nodes that follow in the chain (for instance, in the example the 'Or' function is followed by an 'And'). A linear network, that is a network without any circularities, can be evaluated in a single round of the node list. However, if there are cycles present, more than one evaluation round is required, because the last symbol in a linearized circle is input for the first one, and its value may have changed. Networks may reach a state in which all values are stable, but occasionally, particularly in cases when there is negative feedback in the network, values keep changing forever.
To set up a Boolean simulation, choose the menu option Simulation : Simulation type : Boolean (the default is Continuous). To make the simulation go more slowly, press the Timer button on the Simulation toolbar, and enter 2.0 in the box labelled "Time per evaluation step in seconds" - each evaluation round will now start 2 s after the previous one, rather than the default 0.2 s. Furthermore, tell NetBuilder to automatically stop after a stable state has been reached (in which none of the values change any more) by checking the menu option Simulation : Propagate : Stop when stable (occasionally NetBuilder stops too early when this option is set, so it is always wise to double check whether the state that it has reached is indeed stable).
Make the gene the Probed Symbol, and check that TF1 is on in cell group A, and off in the others (by looking into the other cells). Press Go (the lowest enabled button on the Simulation toolbar), and watch what happens. The document status bar displays the number of evaluation rounds, and you will see that NetBuilder stops automatically after 9 rounds. You can always stop the simulation yourself by hitting the Stop button (underneath Go), which is enabled during a simulation. The simulation may be resumed by pressing Go again.
What has happened? Below the state of the network in cells A, B1, and B3 after evaluation rounds 2, 4, and 6 is displayed (To get these pictures, the diagrams were copied from NetBuilder onto this page via the Windows clipboard):
In rounds 1 and 2, TF1 has switched on the gene in cell group A, and the signal resulting from the activation of this gene has reached the receptor on cell B1 (you specified the intercellular connections earlier, see the page on Receptors). In rounds 4 - 6 the gene in cell B1 is switched on, and the receptors on cells B2 and B3 get activated. In rounds 7 and 8, the gene gets activated in cells B2 and B3, and cell B2 sends a signal back to cell group A, thereby consolidating the activated state of the gene in these cells.
To show that the gene will stay on in cell group A, even when the original activator, TF1, disappears, toggle TF1 off in cell group A, and press Go again. When the network has stabilised again (after one round, which is the round that NetBuilder needs to check that nothing has changed), toggle the inhibitor TF2 on. TF2 is connected to the gene via a 'negated link'. A negated link transfers the negated value of the symbol from which it originates, so that the input (via the gene) to the 'And' function is 1 when TF2 is off, and 0 when TF1 is on. Since the gene will only stay active when TF3 is present and TF2 is absent (or when TF1 is present, regardless of TF2, but we are not concerned with that now) the gene will be switched off in cell group A. Thus, the signal to B1 is lost, and the gene switches off in B1, and eventually also in both other cells.
