Tmc {sdef} | R Documentation |
The function uses Monte Carlo permutations to calculate the empirical distribution of max T(q)=T(qmax) under the null hypothesis of independence among the experiments. An empirical p-value is calculated to evaluate where T(qmax) is located under the null distribution.
Tmc(iter = 1000, output.ratio)
iter |
Number of iteration to be performed |
output.ratio |
The output object from the ratio function |
This function uses Monte Carlo permutations to calculate the empirical distribution of the maximum of T(q) (i.e. T(qmax)) under the null hypothesis of independence among the experiments. While the p-values* for the first list are fixed, the ones for the other lists are independently permutate B times. In this way, any relationship among the lists is destroyed. At each permutation b (b varies from 1 to B) a Tb(q) is calculated for each q and a maximum statistic Tb(qmax) is returned; its distribution represents the null distribution of T(qmax) under the condition of independence. The relative frequency of Tb(qmax) larger than T(qmax) identifies the p-value: it returns the proportion of Tb(qmax) from permuted dataset greater than the observed one (so indicates where the observed T(qmax) is located under the null distribution).
* instead of the p-values any other measure used to rank the features in the experiments can be used
Returns the empirical pvalue from testing T(qmax) and a plot of the Tb(qmax) distribution.
Alberto Cassese, Marta Blangiardo
Stone et al.(1988), Investigations of excess environmental risks around putative sources: statistical problems and a proposed test,Statistics in Medicine, 7, 649-660.
M.Blangiardo and S.Richardson (2007) Statistical tools for synthesizing lists of differentially expressed features in related experiments, Genome Biology, 8, R54.
data = simulation(n=500,GammaA=1,GammaB=1, r1=0.5,r2=0.8,DEfirst=300,DEsecond=200, DEcommon=100) Tq<- ratio(data=data$Pval) MC<- Tmc(iter=100,output.ratio=Tq)