DEResult {puma} | R Documentation |
Class to contain and describe results of a differential expression (DE) analysis. The main components are statistic
which hold the results of any statistic (e.g. p-values, PPLR values, etc.), and FC
which hold the fold changes.
DEResult
objects will generally be created using one of the functions pumaDE
, calculateLimma
, calculateFC
or calculateTtest
.
Objects can also be created from scratch:
new("DEResult")
new("DEResult",
statistic=matrix()
, FC=matrix()
, statisticDescription="unknown"
, DEMethod="unknown"
)
statistic
:FC
:statisticDescription
:statistic
slot.DEMethod
:Class-specific methods.
statistic(DEResult)
, statistic(DEResult,matrix)<-
statistic
slot.FC(DEResult)
, FC(DEResult,matrix)<-
FC
slot.statisticDescription(DEResult)
, statisticDescription(DEResult,character)<-
statisticDescription
slot.DEMethod(DEResult)
, DEMethod(DEResult,character)<-
DEMethod
slot.pLikeValues(object, contrast=1, direction="either")
DEResult
, converted to "p-like values". If the object holds information on more than one contrast, only the values of the statistic for contrast number contrast
are given. Direction can be "either" (meaning we want order genes by probability of being either up- or down-regulated), "up" (meaning we want to order genes by probability of being up-regulated), or "down" (meaning we want to order genes by probability of being down-regulated). "p-like values" are defined as values between 0 and 1, where 0 identifies the highest probability of being differentially expressed, and 1 identifies the lowest probability of being differentially expressed. We use this so that we can easily compare results from methods that provide true p-values (e.g. calculateLimma
) and methods methods that do not provide p-values (e.g. pumaDE
). For objects created using pumaDE
, this returns 1-PPLR if the direction is "up", PPLR if direction is "down", and 1-abs(2*(PPLR-0.5)) if direction is "either". For objects created using calculateLimma
or calculateTtest
, this returns the p-value if direction is "either", ((p-1 * sign(FC))/2)+ 0.5, if the direction is "up", and ((1-p * sign(FC))/2)+ 0.5 if the direction is "down". For all other methods, this returns the rank of the appropriate statistic, scaled to lie between 0 and 1. contrast
will be returned.topGenes(object, numberOfGenes=1, contrast=1, direction="either")
numberOfGenes
specifies the number of genes to be returned by the function. If the object holds information on more than one contrast, only the values of the statistic for contrast number contrast
are given. Direction can be "either" (meaning we want order genes by probability of being either up- or down-regulated), "up" (meaning we want to order genes by probability of being up-ragulated), or "down" (meaning we want to order genes by probability of being down-regulated). Note that genes are ordered by "p-like values" (see pLikeValues
). object
is an object of class DEResult
.topGeneIDs(object, numberOfGenes=1, contrast=1, direction="either")
numberOfGenes
specifies the number of genes to be returned by the function. If the object holds information on more than one contrast, only the values of the statistic for contrast number contrast
are given. Direction can be "either" (meaning we want order genes by probability of being either up- or down-regulated), "up" (meaning we want to order genes by probability of being up-ragulated), or "down" (meaning we want to order genes by probability of being down-regulated). Note that genes are ordered by "p-like values" (see pLikeValues
). object
is an object of class DEResult
.numberOfProbesets(object)
DEResult
. This method is synonymous with numberOfGenes.numberOfGenes(object)
DEResult
. This method is synonymous with numberOfProbesets.numberOfContrasts(object)
DEResult
.write.reslts(object)
signature(x = "DEResult")
: writes the statistics and related fold changes (FCs) to
files. It takes the same arguments as write.table
. The argument "file" does not need to set any
extension. The different file marks and extension "csv" will be added automatically. The default file name is "tmp".
In the final results, statistics are in the file "tmp_statistics.csv", and FCs are in
"tmp_FCs.csv" respectively. Standard generic methods:
show(object)
Richard D. Pearson
Related methods pumaDE
, calculateLimma
, calculateFC
or calculateTtest
.
## Create an example DEResult object # Next 4 lines commented out to save time in package checks, and saved version used # if (require(affydata)) { # data(Dilution) # eset_mmgmos <- mmgmos(Dilution) # } data(eset_mmgmos) # Next line used so eset_mmgmos only has information about the liver factor # The scanner factor will thus be ignored, and the two arrays of each level # of the liver factor will be treated as replicates pData(eset_mmgmos) <- pData(eset_mmgmos)[,1,drop=FALSE] # To save time we'll just use 100 probe sets for the example eset_mmgmos_100 <- eset_mmgmos[1:100,] eset_comb <- pumaComb(eset_mmgmos_100) esetDE <- pumaDE(eset_comb) ## Use some of the methods statisticDescription(esetDE) DEMethod(esetDE) numberOfProbesets(esetDE) numberOfContrasts(esetDE) topGenes(esetDE) topGenes(esetDE, 3) pLikeValues(esetDE)[topGenes(esetDE,3)] topGeneIDs(esetDE, 3) topGeneIDs(esetDE, 3, direction="down") ## save the expression results into files write.reslts(esetDE, file="example")