Description Usage Arguments Value Author(s) References See Also Examples
This function runs function ARTIVAsubnet
for all target
genes in targetData
successively. This function generates
Reversible Jump MCMC (RJ-MCMC) sampling for approximating the posterior
distribution of a time varying regulatory network, under the Auto
Regressive TIme VArying (ARTIVA) model (for a detailed description of
the algorithm, see ARTIVAsubnet
and see Lebre et al.
BMC Systems Biology, 2010). A network representing the interactions
between the factor genes and the target genes is estimated and plotted.
1 2 3 4 5 6 7 8 | ARTIVAnet(targetData, parentData, targetNames = NULL, parentNames =
NULL, dataDescription=NULL, saveEstimations=TRUE, saveIterations=FALSE,
savePictures = TRUE, outputPath=NULL, dyn=1, segMinLength=2, maxCP=NULL,
maxPred=NULL, nbCPinit=NULL, CPinit=NULL, niter=50000, burn_in=NULL,
PSRFactor=FALSE,PSRF_thres=1.05, segmentAnalysis=TRUE,
edgesThreshold=0.5, layout = "fruchterman.reingold", cCP= 0.5,
cEdges=0.5, alphaCP=1, betaCP=0.5, alphaEdges=1, betaEdges=0.5, v0=1,
gamma0=0.1, alphad2=2, betad2=0.2, silent=FALSE)
|
targetData |
A vector with the temporal gene expression measurements for the target gene (i.e. the gene whose regulation factors are looked for). |
parentData |
A matrix (or a vector if only 1 parent gene) with the temporal gene expression measurements for the proposed parent genes (i.e. potential regulation factors). Parent genes are shown in row and expression values in column. For computational reasons, we advise not to test simultaneously more than 10 parent genes. |
targetNames |
A vector with the names for target gene(s) (optional, default:
|
parentNames |
A vector with the names for parent gene(s) (optional, default:
|
dataDescription |
(Required only when the gene expression measurements contain repeated
values for the same time points). A vector indicating the ordering of
the time measurements in the data. For example
dataDescription=rep(1:n, each=2), if there are two measurements for
each time point AND the repetitions for each time point are next to
each other. Note that temporal gene expression measurements have to be
organized identically in arguments |
saveEstimations |
Boolean, if |
saveIterations |
Boolean, if |
savePictures |
Boolean, if |
outputPath |
File path to a folder in which the output results have to be saved,
either a complete path or the name of a folder to be created in the
current directory (optional, default: |
dyn |
Time delay to be considered in the auto-regressive process, between the
temporal expression measurements of the analyzed target gene and the
ones of the parent genes (optional, default: |
segMinLength |
Minimal length (number of time points) to define a temporal
segment. Must be - strictly - greater than 1 if there is no
repeated measurements for each time point in arguments
|
maxCP |
Maximal number of CPs to be considered by the ARTIVA inference
procedure. Note that for a temporal course with |
maxPred |
Maximal number of simultaneous incoming edges for each segment of the
regulatory network estimated for the target gene (default:
|
nbCPinit |
Number of CPs to be considered at the algorithm initialization
(optional, default: |
CPinit |
A vector with the initial positions for potential CPs.
(optional, default: |
niter |
Number of iterations to be performed in the RJ-MCMC sampling
(optional, default: |
burn_in |
Number of initial iterations that are discarded for the estimation of
the model distribution (posterior distribution). The
|
PSRFactor |
Boolean, if |
PSRF_thres |
(Only when |
segmentAnalysis |
Boolean, if |
edgesThreshold |
Probability threshold for the selection of the edges of the time-varying
regulatory network when |
layout |
Name of the function determining the placement of the vertices for
drawing a graph, possible values among others:
|
cCP |
Maximal probability to propose the birth (resp. death) of a changepoint
(CP) during the RJ-MCMC iterations (optional, default: |
cEdges |
Maximal probability - when a move update of the network topology is
chosen - to propose ecah edge move (birth or death of an edge) within a
temporal segment (optional, default: |
alphaCP |
Hyperparameter for sampling the number |
betaCP |
Hyperparameter for sampling the number |
alphaEdges |
Hyperparameter for sampling the number |
betaEdges |
Hyperparameter for sampling the number |
v0 |
Hyperparameter for sampling the noise variance (denoted by |
gamma0 |
Hyperparameter for sampling the noise variance (denoted by |
alphad2 |
Hyperparameter for sampling a parameter that represents the expected
signal-to-noise ratio (denoted by |
betad2 |
Hyperparameter for sampling a parameter that represents the expected
signal-to-noise ratio (denoted by |
silent |
Boolean, if |
A table containing the information to plot (see function
traceNetworks
) the global network estimated by the
ARTIVAnet
procedure.
All results are plotted in a pdf file (when choosing savePictures =
TRUE
) in folder outputPath
.
All numerical results (see ARTIVAsubnet
output values
documentation)) are saved in text files (when choosing saveEstimations=TRUE
and/or saveIterations=TRUE
) in folder outputPath
.
S. Lebre and G. Lelandais
Statistical inference of the time-varying structure of gene-regulation networks S. Lebre, J. Becq, F. Devaux, M. P. H. Stumpf, G. Lelandais, BMC Systems Biology 2010, 4:130.
Inference from iterative simulation using multiple sequences. Gelman, A. and D. Rubin, Statistical science 7 (4), 457-472, 1992.
ARTIVAsubnet
, choosePriors
,
ARTIVAsubnetAnalysis
,CP.postDist
,
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 | library(ARTIVA)
# Load the dataset with simulated gene expression profiles
data(simulatedProfiles)
# List of target genes to be analyzed independantly with ARTIVA
targetGenes = c(1, 10, 20, "TF3", 45, 50)
# Names of the parent genes (typically transcription factors)
parentGenes = c("TF1", "TF2", "TF3", "TF4", "TF5")
###
# ARTIVA analysis searching for potential interactions between each target
# genes and a predefined list of parent genes.
###
# Note that the number of iterations in the RJ-MCMC sampling is reduced
# to 'niter=20000' in this example, but it should be increased (e.g. up to
# 50000) for a better estimation.
## Not run:
ARTIVAtest1=ARTIVAnet(targetData = simulatedProfiles[targetGenes,],
parentData = simulatedProfiles[parentGenes,],
targetNames= targetGenes,
parentNames = parentGenes,
niter = 20000,
savePictures=FALSE)
# Note that function ARTIVAnet calls fonction ARTIVAsubnet for each
# target gene successively and provides a global estimated regulatory
# network entitled "ARTIVA_FinalNetwork.pdf" in addition to the output
# results given by function ARTIVAsubnet.
## Gene names for the target and the parent genes, minimum segment length,
## threshold for the edges selection can be specified as follow:
ARTIVAtest2=ARTIVAnet(targetData = simulatedProfiles[targetGenes,],
parentData = simulatedProfiles[parentGenes,],
targetNames= targetGenes,
parentNames = parentGenes,
segMinLength = 2,
edgesThreshold = 0.5,
niter = 20000,
outputPath = "ARTIVA-test")
## End(Not run)
# By default, the output results (pictures and estimation values) are
# saved in a folder named "ARTIVAnet" created in the current directory
# In order to save the results in a specific folder, for example a
# folder entitled "ARTIVA-test" in the current directory:
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.