mLL {nem} | R Documentation |
computes the marginal likelihood of observed phenotypic data given a phenotypic hierarchy.
mLL(Phi,D1,D0=NULL,a=0.15,b=0.05,Pe=NULL,Pm=NULL,lambda=0,type="mLL")
Phi |
an adjacency matrix with unit main diagonal |
D1 |
(i) count matrix for discrete data: phenotypes x genes. How often did we see an effect after interventions? (ii) matrix describing the probabilities of an effect (iii) probability density matrix discribing the strength of an effect (iv) log-odds ratio matrix describing the strength of an effect |
D0 |
count matrix: phenotypes x genes. How often did we NOT see an effect after intervention? Not used for continious data |
a |
false positive rate: how probable is it to miss an effect? (for count matrix) |
b |
false negative rate: how probable is it to see a spurious effect? (for count matrix) |
Pe |
prior of effect reporter positions in the phenotypic hierarchy |
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. |
type |
(i) "mLL" = marginal likelihood using count matrices (ii) "CONTmLL" = marginal likelihood for probability matrices (iii) "CONTmLLDens" = marginal likelihood for probability density matrices |
It computes the marginal likelihood of a single phenotypic hierarchy.
Usually called from within the function score
.
mLL |
marginal likelihood of a phenotypic hierarchy |
pos |
posterior distribution of effect positions in the hierarchy |
mappos |
Maximum aposteriori estimate of effect positions |
Florian Markowetz <URL: http://genomics.princeton.edu/~florian>
Markowetz F, Bloch J, Spang R, Non-transcriptional pathway features reconstructed from secondary effects of RNA interference, Bioinformatics, 2005
data("BoutrosRNAi2002") result <- nem(BoutrosRNAiDiscrete[,9:16],type="mLL",para=c(.15,.05))