BNMPy.PBN

class BNMPy.PBN.ProbabilisticBN(numberOfNodes, linkages, numberOfFunctions, functions, probabilities, initialNodeValues, outputFilePath='', nodeDict=None)

Methods

buildK()	This rebuilds the K array and related attributes.
copy()	Create a deep copy of the PBN.
knockdown(key, value[, efficacy])	Sets a specific node with partial efficacy for experimental perturbation.
knockout(key, value)	Sets a specific node to be permanently fixed to a given value.
setInitialValue(key, value)	Sets a particular node to a given initial value, where the key is indexed in nodeDict.
setInitialValues(initialNodeValues)	Sets the initial values of the probabilistic boolean network.
undoKnockouts()	Undoes all knockouts and knockdowns.
update([iterations])	Update the network directly.
update_cumulative_probabilities()	Update cumulative probabilities for each node's functions.
update_noise(p[, iterations])	Update the network with noise parameter p over a given number of iterations.
update_one()	Asynchronous update: Update one random non-constant node for one step.

getBias
getMaxConnectivity
getMeanConnectivity
getRealization
getTrajectory
initializeOutput
stateToWrite
writeNetworkHistory

buildK()

This rebuilds the K array and related attributes.

copy() → ProbabilisticBN

Create a deep copy of the PBN. Required for optimization to avoid modifying original network.

getBias()

getMaxConnectivity()

getMeanConnectivity()

getRealization()

getTrajectory()

initializeOutput()

knockdown(key, value, efficacy=1.0)

Sets a specific node with partial efficacy for experimental perturbation. E.g., when using a drug to inhibit a gene, the efficacy is the probability of the drug being effective.

knockout(key, value)

Sets a specific node to be permanently fixed to a given value.

setInitialValue(key, value)

Sets a particular node to a given initial value, where the key is indexed in nodeDict.

setInitialValues(initialNodeValues)

Sets the initial values of the probabilistic boolean network.

stateToWrite()

undoKnockouts()

Undoes all knockouts and knockdowns. Does not change initial values, however.

update(iterations=1)

Update the network directly. Select a function for each node based on its probability distribution. Then, update the network state based on the selected function.

update_cumulative_probabilities()

Update cumulative probabilities for each node’s functions. This is used for efficient function selection during simulation.

update_noise(p, iterations=1)

Update the network with noise parameter p over a given number of iterations.

This method simulates the network’s evolution over time with added noise. Noise randomly flips node states with probability p after normal network update, but is never applied to constant nodes (knockouts).

update_one()

Asynchronous update: Update one random non-constant node for one step.

This method randomly selects one non-constant node and updates only that node based on its Boolean function, while keeping all other nodes unchanged. This simulates asynchronous dynamics where nodes update at different times.

writeNetworkHistory()