Description Usage Arguments Value Examples
Main Function for Individual Study DE: microarray & RNAseq
The Indi.DE.Analysis
is a function to perform individual association
analysis between gene expression and the response/phenoype of interest (can
be either group, continuous or survival).
1 2 3 4 |
data |
is a list of K elements, where K is the number of studies, each element is a microarray or RNAseq expression matrix with G rows and N columns, where G is number of matched genes and N is the sample size. |
clin.data |
is a list of K elements, each element includes is a clinical data frame with N rows and p columns, where N is the sample size and p is the number of clinical variables (main response included). |
data.type |
is a character indicating the data type of the elements
in |
resp.type |
is a character indicating the response type of the
|
response |
is one column name of |
covariate |
are the clinical covariates you wish to adjust for in the DE analysis, can be a vector of column names or NULL. |
ind.method |
is a character vector to specify the method used to test if there is association between the gene expression and outcome variable. must be one of "limma", "sam" for "continuous" data type and "edgeR", "DESeq2" or "limmaVoom" for "discrete" data type. |
select.group: |
for two-class comparison only, specify the two groups for comparison when the group factor has more than two levels. |
ref.level: |
for two-class/multi-class comparison only, specify the reference level of the group factor. |
paired: |
logical value indicating whether paired design; |
asymptotic: |
a logical value indicating whether asymptotic distribution should be used. If FALSE, permutation will be performed. |
nperm: |
the number of permutations. Applicable when |
tail: |
a character string specifying the alternative hypothesis, must be one of "abs" (default), "low" or "high". For resp.type = "continuous", "survival" only. |
seed: |
Optional initial seed for random number generator. |
a list with components:
p: For all types of response, the p-value of the association test for each gene
stat: For "continuous" and "survival" only, the value of test statistic for each ##' gene
bp: For "continuous" and "survival" only, the p-value from
nperm:
permutations for each gene. It will be used for the meta
analysis by default. It can be NULL if you chose asymptotic results.
log2FC: For "twoclass" only, the log2 fold change for each gene
lfcSE: For "twoclass" only, the standard error of log2 fold change for each gene
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 | data('Leukemia')
data('LeukemiaLabel')
data <- Leukemia
K <- length(data)
clin.data <- lapply(label, function(x) {data.frame(x)} )
for (k in 1:length(clin.data)){
colnames(clin.data[[k]]) <- "label"
}
select.group <- c('inv(16)','t(15;17)')
ref.level <- "inv(16)"
data.type <- "continuous"
ind.method <- c('limma','limma','limma')
resp.type <- "twoclass"
paired <- rep(FALSE,length(data))
ind.res <- Indi.DE.Analysis(data=data,clin.data= clin.data,
data.type=data.type,resp.type = resp.type,
response='label',
ind.method=ind.method,select.group = select.group,
ref.level=ref.level,paired=paired)
N <- sapply(data, FUN=function(x) ncol(x))
survival.time <- lapply(N,FUN = function(x) round(runif(x,10,2000)))
censor.status <- lapply(N,FUN = function(x) sample(c(0,1),x,replace=TRUE) )
for (k in 1:length(clin.data)){
clin.data[[k]] <- cbind(clin.data[[k]],survival.time[[k]],censor.status[[k]])
colnames(clin.data[[k]])[2:3] <- c("survival","censor")
}
ind.method <- c('logrank','logrank','logrank')
resp.type <- "survival"
ind.res <- Indi.DE.Analysis(data=data,clin.data= clin.data,
data.type=data.type,resp.type = resp.type,
response=c("survival","censor"),
ind.method=ind.method,asymptotic=TRUE)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.