ab.mra: Inference of the effect of a double perturbation upon the...

In bioinfo-ircm/aiMeRA: Modular Response Analysis for the quantitative analysis of connectivity in biological networks

Description

This function first calculates the coupling between the elements of a biological network using MRA and then, it infers the output of combining two of the perturbations used for the construction of the model.

Usage

ab.mra(data,matp,pred=NULL,pert1,pert2,inval=c(-1,1),step=0.1,Rp=FALSE,ab=TRUE)

Arguments

data	A data frame containing the experimental data in the specific format for MRA calculations.
matp	The perturbation matrix. Names of modules (rows) and perturbations (columns) must correspond to names of rows and columns in tab.
pred	String. Name of the double perturbation to be inferred (only if the experimental values for all modules of such perturbation are given as reference in “'data“'). Default is NULL.
pert1	String. Name of the first individual peturbation.
pert2	String. Name of the second individual perturbation.
inval	A two values vector given the lower and the upper limit of the interval of the a and b coefficients for the ab.mra calculation (See details).
step	Number. Increment of the sequence of the interval for the a and b coefficients.
Rp	Logical. TRUE if “'data“' is the calcuated global response matrix. Default is FALSE
ab	Logical. If TRUE then the inferred values of a double perturbation are obtained by using the a and b coefficients as defined for the ab.mra calculation. If ab=FALSE then a=1 and b=1 (See details).

Details

The ab.MRA inference is based on the hypothesis that it is possible to infer experimental values of combining two perturbations used for the classical MRA calculation network coonectivity. The hypothesis is that by including the two local responses of two perturbations into a new system for which the network connectivity is known then it is possible to infer the values of a double perturbation effet upon all biological modules in the network.

Two coefficients (a and b) that ponderate the local responses to perturbations (local_matrix) indicate whether the effect of the combined perturbations remains equal as their individual effect (a=1 and b=1), or it is attenuated (abs(a)<1 and abs(b)<1) or amplified (abs(a)>1 and abs(b)>1). The a, b coefficients are coefficients that minimize the euclidian distance between the infered values and the reference experimental values of all modules.

Value

List. If “'pert“' is provided and “'ab=TRUE“' then the inferred value, the refrence data and the values for the a and b coefficients are returned. If “'pert“' is provided and “'ab=FALSE“' then the inferred and the reference values are returned. If “'pert“' is not provided then only the inferred values are returned. This will produce a warning message suggesting that the inferred values must be validated by experimental data.

Examples

#Inferrence of a double perturbation by two siRNAs (siRIP140 and siLCoR)
#in a E2 stimulated biological condition of the ERa-RIP140-LCoR network.
data=data.setup(list(estr1_A,estr1_B,estr2_A,estr2_B,estr3_A,estr3_B))
data.mean=data2sdmean(data)$mean
rules=c("Et->Luciferase","E2+siRIP140->RIP140","E2+siLCoR->LCoR","E2->0")
matp=read.rules(rules)
#Inferrence with the experimental value of reference and without the a and b coefficients
ab.mra(data.mean,matp=matp,pred="E2+siLCoR+siRIP140",pert1="E2+siLCoR",
       pert2="E2+siRIP140",ab=FALSE)
#Inferrence with the experimental value of reference and the a and b coefficients
ab.mra(data.mean,matp=matp,pred="E2+siLCoR+siRIP140",pert1="E2+siLCoR",
       pert2="E2+siRIP140")
#Inferrence without the expeimental value of reference and without the a and b coefficients.
ab.mra(data.mean,matp=matp,pred="E2+siLCoR+siRIP140",pert1="E2+siLCoR",
       pert2="E2+siRIP140")

#Inferrence of a biological module (GREB1) without the a,b coefficients,
#which does not have an individual perturbation
data=data.setup(list(estr1_A,estr1_B,estr2_A,estr2_B,estr3_A,estr3_B))
data.mean=data2sdmean(data)$mean
rules=c("Et->Luciferase","E2+siRIP140->RIP140","E2+siLCoR->LCoR","E2->0","0->GREB1")
ab.mra(data.mean,matp=matp,pred="E2+siLCoR+siRIP140",pert1="E2+siLCoR",pert2="E2+siRIP140",
      ab=FALSE)

bioinfo-ircm/aiMeRA documentation built on Oct. 1, 2021, 8:11 a.m.