Calculates the distribution of overlaps between query and arbitrary genomic partitions

Takes a GRanges object, then assigns each element to a partition from the provided partitionList, and then tallies the number of regions assigned to each partition. A typical example of partitions is promoter, exon, intron, etc; this function will yield the number of each for a query GRanges object There will be a priority order to these, to account for regions that may overlap multiple genomic partitions.

calcPartitions(
  query,
  partitionList,
  remainder = "intergenic",
  bpProportion = FALSE
)

Arguments

query

GRanges or GRangesList with regions to classify

partitionList

an ORDERED (if bpProportion=FALSE) and NAMED list of genomic partitions GRanges. This list must be in priority order; the input will be assigned to the first partition it overlaps. bpProportion=TRUE, the list does not need ordering.

remainder

A character vector to assign any query regions that do not overlap with anything in the partitionList. Defaults to "intergenic"

bpProportion

logical indicating if overlaps should be calculated based on number of base pairs overlapping with each partition. bpProportion=FALSE does overlaps in priority order, bpProportion=TRUE counts number of overlapping base pairs between query and each partition.

Value

A data.frame assigning each element of a GRanges object to a partition from a previously provided partitionList.

Examples

partitionList = genomePartitionList(geneModels_hg19$genesGR,
                                    geneModels_hg19$exonsGR,
                                    geneModels_hg19$threeUTRGR,
                                    geneModels_hg19$fiveUTRGR)
calcPartitions(vistaEnhancers, partitionList)
#>      partition Freq
#> 1         exon   30
#> 2      fiveUTR   11
#> 3   intergenic  828
#> 4       intron  441
#> 5 promoterCore    5
#> 6 promoterProx   12
#> 7     threeUTR   12