Get_Attractor_parallel.f: Get the attractor given an initial condition.
In SPIDDOR/SPIDDOR: SPIDDOR

Description Usage Arguments Details Value References See Also

Get the attractor given an initial condition using a parallelized code using the snowfall package.

Get_Attractor_parallel.f(cpus,
                        BN,
                        time.steps=999,
                        Knockouts="",  
                        Over_expr="", 
                        Over_expr_AA="",
                        KO_times=NULL,
                        OE_times=NULL,
                        asynchronous=TRUE,
                        repetitions,
                        Percent.ON=TRUE)

`cpus`	Number of CPUs requested for the cluster in order to parallelize the code
`BN`	A list containing the most relevant information of the loaded network. See `read.Boolean.functions` for more details.
`time.steps`	A number specifying the iterations for the simulation algorithm. We recommend to use the default value of 999 (1000 time steps in total, including the initial conditions).
`Knockouts`	A character vector with the name of the nodes to knockout (fixed to 0) over all the simulation (empty by default).
`Over_expr`	A character vector with the name of the nodes to overexpress (fixed to 1) over all the simulation (empty by default).
`Over_expr_AA`	A character vector with the name of the nodes to overexpress (fixed to 1) after their first activation (empty by default).
`KO_times`	A numeric vector specifying the iterations where the nodes in Knockouts argument will be fixed to 0. If empty the knockout is applied to the nodes for the entire simulation (empty by default).
`OE_times`	A numeric vector specifying the iterations where the nodes in Over_expr argument will be fixed to 1. If empty the overexpression is applied to the nodes for the entire simulation (empty by default).
`asynchronous`	logical; if FALSE the synchronous mode is selected; if TRUE the asynchronous scheme is chosen (by default).
`repetitions`	Number of repetitions to re-compute the simulation algorithm in order to get the attractor. We recommend first to try with a number smaller that 20. More than 100 repetitions are not allowed.
`Percent.ON`	logical. If TRUE returns the attractor represented as the probability of activation of the nodes in the network. If FALSE returns all the states that form the attractor. Default is TRUE.

The initial conditions of the network can be specified in the section Initial_conditions of the BN list (BN$Initial_conditions).

If synchronous mode is selected, the attractor could be a fixed point or a simple/complex cycle. If asynchronous method is prefered, the resulting attractor could be a fixed point or a complex attrator. Complex attractors consist on set of states in which the system irregularly oscillates, making its biological interpretation quite difficult. If Percent.ON=TRUE is choosen, this function generates the probability that a given node is in ON state inside the complex attractor. This allows users to overcome this interpretation barrier.

If a large network is being analyze the simulation algorithm must be repeated several times to get the proper attractor of the network.

If Percent.ON=TRUE is selected, the function returns the attractor represented with the probability of activation of the nodes. If Percent.ON=FALSE is choosen, it returns all the states that form the attractor in a data.frame.

Saadatpour A, Albert I, Albert R (2010). Attractor analysis of asynchronous Boolean models of signal transduction networks. J Theor Biol.

sfInit sfClusterApplyLB sfStop

SPIDDOR/SPIDDOR documentation built on May 9, 2019, 11:42 a.m.