defineBins: Define Bins

Description

Defines bins about the 5' end of certain features of interest - these features are usually transcripts.

Usage

defineBins(anno, zone, geneID="ensembl_gene_id", removeDuplicates=TRUE)

Arguments

anno

data.frame (or, an object that can be coerced to a data.frame, such as a GRanges) - Annotation information, corresponding to features of interest (usually transcripts). Only the 5' end of each object is used.

Rcade expects the following column names: chr, start, end, str. These correspond to chromosome name, start co-ordinate, end co-ordinate and strand. Additionally, there must be another column specifying a gene ID, specified by the geneID argument.

zone

integer - must be a length 2 vector of form c(relative.start, relative.end). For example, zone = c(-10,100) will produce bins that start 10bp 5' of each transcript's TSS and end 100bp 3' of it.

geneID

character or integer - The column in anno that contains a geneID (or some other feature ID).

removeDuplicates

logical - If TRUE, then any rows that share the same geneID and genomic location as another row will be removed (even if any of the other columns are different).

Value

A GRanges object, corresponding to genomic bins. This output can be used as the ChIPannoZones argument in RcadeAnalysis.

Details

The defineBins function is useful when ChIP-seq bins are defined about ... . In particular, biomaRt data can be fed into this function directly. FIXME See vignette.

Examples

Run this code



## Not run: 	##acquire annotation from biomaRt
# 	library(biomaRt)
# 	anno <- getBM(
# 		attributes= c("ensembl_gene_id", "chromosome_name",
# 			"transcript_start", "transcript_end", "strand"),
# 		mart= useDataset("hsapiens_gene_ensembl", useMart("ensembl"))
# 	)
# ## End(Not run)
#define bins about the annotation
anno <- anno[order(anno$chromosome_name),]
colnames(anno) <- c("ENSG","chr","start","end","str")
ChIPannoZones <- defineBins(anno, c(-1500, 1500), geneID = "ENSG")