genomic_corr_reldist: Relative distance between two sets of genomic regions
In jokergoo/epik: Integrative Analysis and Visualization of Epigenomic Sequencing Data

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

View source: R/genomic_region_correlation.R

Relative distance between two sets of genomic regions

1	genomic_corr_reldist(gr1, gr2)

`gr1`	genomic region 1, a `GRanges` object
`gr2`	genomic region 2, a `GRanges` object

For regions in gr1 and gr2, they are all degenerated as single points which are the middle points of regions. For each middle point in gr1, it looks for two nearest points in gr2 on its left and right. The statistic is defined as the ratio of the distance to the nearest neighbour point to the distance of two neighbour points. If gr1 and gr2 are not correlated at all, It is expected that the ratio follows a uniform distribution. So final statisitic are the KS-statistic between the real distribution of rations to the uniform distribution.

A single correlation value.

Zuguang Gu <z.gu@dkfz.de>

Favoriv A, et al. Exploring massive, genome scale datasets with the GenometriCorr package. PLoS Comput Biol. 2012 May; 8(5):e1002529

genomic_regions_correlation

1
2
3

gr1 = GRanges(seqnames = "chr1", ranges = IRanges(c(1, 5), c(3, 8)))
gr2 = GRanges(seqnames = "chr1", ranges = IRanges(c(2, 6), c(4, 8)))
genomic_corr_reldist(gr1, gr2)