We are now done with the specification of the reactions in our model.

To make the model ready for use in simulations, we have to make sure that we have all those complexes defined that we want to use to initialize the biochemistry of the simulated cells or for tracking the evolution of the reaction network over time.

At this point, we should briefly discuss what makes a complex fit to serve as a component for initial biochemical states.

In this tutorial, we have tried to carefully distinguish between complex species and complexes. Complex species possess features such as binding sites, intra- or extracellular components and (optionally) state features, such as the GTP tag we specified for the Galpha species. Complexes are particular instances of such species where features are set to a specific state: for the auto-GTPase activity of Galpha, for instance, we defined an initial complex where GTP was 'on' and a result complex where this feature was 'off'.

Complexes where all binding site states (which can be bound/unbound/don't care) and all features states are specified without ambiguity are called 'well-defined' complexes. For the feature states this means they have to be 'on' or 'off' but not 'don't care'. For the binding site states this means, they all have to be 'unbound'. Setting the binding state of a binding site to 'bound' without providing a specific binding partner introduces ambiguity with regard to the binding partner. This may be wanted for some reaction rules -- if we want to express that a certain reaction can only happen after a certain site has been bound (regardless of what bound to it).

Patterns are also very useful to describe efficiently several complexes of the same species that share a specific feature (for instance all complexes of the Galpha species that have the GTP tag set to 'on'). During a simulation, querying the concentration of a pattern will yield the sum of the concentration of all complexes matching the pattern.

As soon as a complex contains feature states set to 'don't care' or binding sites not set to 'unbound' that complex is called a 'pattern' since it potentially matches several well-defined complexes.

Complexes that are meant to specify initial biochemical states (which molecule complexes are there and what are their initial concentrations?) must be well-defined since concentrations cannot be set without knowing for which complexes, exactly. So we need to make sure we have the needed well-defined initial complex states.

We will start with the basal state of the G proteins. Load the Galpha.Gbetagamma complex species into the Complex Species view by double-clicking it in the list on the left. Then, go to Relations->New and select 'Create/Edit Complex'.

The Modeler will, as a default, suggest an (externally) unbound complex with all feature tags set to 'off'. In this case, this is exactly the complex we want to use as a basal state for G protein complex. So, we simply adjust the name to reflect this (set it to GalphaGDP_Gbetagamma_unbnd, for example. Or just Gabg_basal.). Don't forget to click 'Save'.

Similarly, we will define a basal state for the Receptor.

The suggested complex is, indeed, unbound and we just need to select a more descriptive name again and save it.

To keep track of the total concentration of activated Galpha (i.e. in the GTP 'on') state we can define a pattern "Galpha_GTP_all" where the binding states of the two binding sites are set to 'don't care'. This pattern will then cover all complexes containing the activated Galpha molecule.

...and a pattern that keeps track of the complement: "Galpha_GDP_all".

The file "G_protein_7.6.dbf" represents the model as it should be at this stage of the tutorial.