Computing Boolean networks dynamics with minibn

The module minibn implements update modes for computing dynamics of minibn.BooleanNetwork objects. It is returned as networkx.DiGraph object.

from colomoto import minibn

Let us first define a simple Boolean network.

bn = minibn.BooleanNetwork({
    "a": "c",
    "b": "!a",
    "c": "!b"
})

By default the .dynamics method computes the dynamics according to the (fully) asynchronous update mode.

adyn = bn.dynamics()
adyn

# computing graph layout...

The networkx module implements a wide range of graph algorithms, such as computing strongly connected components:

import networkx as nx

list(nx.strongly_connected_components(adyn))

[{'010'}, {'101'}, {'000', '001', '011', '100', '110', '111'}]

The update mode can be specified as first argument. It can be either "asynchronous" (or equivalently "fully-asynchronous"), "general", or "synchronous". Custom update modes can also be implemented by inheriting from the minibn.UpdateModeDynamics.

bn.dynamics("synchronous")

# computing graph layout...

Finally, one can also compute the dynamics only reachable from a specified initial state, specified with the init keyword:

bn.dynamics("synchronous", init={"a": 1, "b": 1, "c": 1})

# computing graph layout...