annotFilter: Checks and filters an annotation list.
In mdgsa: Multi Dimensional Gene Set Analysis.

Description Usage Arguments Details Value Author(s) See Also Examples

Checks that the annotated genes (those in the annotation list) are consistent with the universe of genes defined by the ranking index. Filters out functional blocks too 'big' or too 'small'.

1 2	annotFilter(annot, index, minBlockSize = 10, maxBlockSize = 500, verbose = TRUE)

`annot`	an annotation list.
`index`	ranking index. Vector, matrix or data.frame
`minBlockSize`	minimum block size kept
`maxBlockSize`	maximum block size kept
`verbose`	verbose

index is optional. When it is not provided, the annotation lists is just filtered out by the sizes of the blocks of genes defined in the list.

If a ranking index is provided its names are assumed to be the universe of genes under study. The genes in the annotation list are compared against those of the ranking index and the ones not belonging to the universe are removed out form the annotation in order to compute the size of each functional block. Then the list is filtered by sizes; too big and too small blocks are removed.

No transformation is done over the ranking index or its names (gene IDs).

index may just be a character vector containing the names of the genes in the universe, that is, the names or row names of the ranking index.

a filtered annotation list.

David Montaner dmontaner@cipf.es

annotMat2list

rindex <- 1:10
names (rindex) <- paste ("gene", rindex, sep = "")
rindex

annot <- list (paste ("gene", 1:2, sep = ""),          ##too small block
               paste ("gene", 1:6, sep = ""),          ##right size
               paste ("gene", 1:5, sep = ""),          ##too big block
               paste ("gene", c(1:3, 1:3), sep = ""))  ##duplicated IDs
annot[[2]][1] <- NA
annot[[2]][2] <- ""
annot[[2]][3] <- "BAD_ID"
annot

annotFilter (annot, minBlockSize = 3, maxBlockSize = 5)
annotFilter (annot, rindex, minBlockSize = 3, maxBlockSize = 5)