triples.posterior {nem} | R Documentation |
Function triples.posterior
estimates the hierarchy triple-wise. In each step only a triple of nodes
is involved and no exhaustive enumeration of model space is needed as in function score
.
triples.posterior(D, type="mLL",para=NULL, hyperpara=NULL,Pe=NULL,Pmlocal=NULL,Pm=NULL,lambda=0, triples.thrsh=.5, selEGenes=FALSE, verbose=TRUE) #S3 methods for class 'triples' print.triples(x,...)
D |
data matrix. Columns correspond to the nodes in the silencing scheme. Rows are phenotypes. |
type |
(1.) marginal likelihood "mLL" (only for cout matrix D), or (2.) full marginal likelihood "FULLmLL" integrated over a and b and depending on hyperparameters a0, a1, b0, b1 (only for count matrix D), or (3.) "CONTmLL" marginal likelihood for probability matrices, or (4.) "CONTmLLDens" marginal likelihood for probability density matrices, or (5.) "CONTmLLRatio" for log-odds ratio matrices |
para |
vector with parameters a and b for "mLL", if count matrices are used |
hyperpara |
vector with hyperparameters a0, b0, a1, b1 for "FULLmLL" |
Pe |
prior position of effect reporters. Default: uniform over nodes in hierarchy |
Pmlocal |
local model prior for the four models tested at each node: a vector of length 4 with positive entries summing to one |
triples.thrsh |
threshold used when combining tripel models for each edge. Default: only edges appearing in more than half of triples are included in the final graph. |
Pm |
prior on model graph (n x n matrix) with entries 0 <= priorPhi[i,j] <= 1 describing the probability of an edge between gene i and gene j. |
lambda |
regularization parameter to incorporate prior assumptions. |
selEGenes |
optimize selection of E-genes for each model |
verbose |
do you want to see progress statements printed or not? Default: TRUE |
x |
nem object |
... |
other arguments to pass |
triples.posterior
is an alternative to exhaustive search
by the function score
and more accurate than pairwise.posterior
.
For each triple of perturbed genes
it chooses between the 29 possible models. It then uses model averaging to combine the triple-models into a final graph.
print.triples
gives an overview over the 'triples' object.
graph |
the inferred directed graph (graphNEL object) |
avg |
matrix of edge frequencies in triple models |
pos |
posterior over effect positions |
mappos |
MAP estimate of effect positions |
type |
as used in function call |
para |
as used in function call |
hyperpara |
as used in function call |
lambda |
as in function call |
Florian Markowetz <URL: http://genomics.princeton.edu/~florian>
Markowetz F, Kostka D, Troyanskaya OG, Spang R: Nested effects models for high-dimensional phenotyping screens. Bioinformatics. 2007; 23(13):i305-12.
data("BoutrosRNAi2002") res <- triples.posterior(BoutrosRNAiDiscrete[,9:16],para=c(.13,.05)) # plot graph plot(res,what="graph") # plot posterior over effect positions plot(res,what="pos") # estimate of effect positions res$mappos