Quantifies the extent of overlap between to sets of intervals in terms of base-pairs. Groups that are shared between input are used to calculate the statistic for subsets of data.
Value
tibble with the following columns:
len_ilength of the intersection in base-pairslen_ulength of the union in base-pairsjaccardvalue of jaccard statisticn_intnumber of intersecting intervals betweenxandy
If inputs are grouped, the return value will contain one set of values per group.
Details
The Jaccard statistic takes values of [0,1] and is measured as:
$$ J(x,y) = \frac{\mid x \bigcap y \mid} {\mid x \bigcup y \mid} = \frac{\mid x \bigcap y \mid} {\mid x \mid + \mid y \mid - \mid x \bigcap y \mid} $$
Interval statistics can be used in combination with
dplyr::group_by() and dplyr::do() to calculate
statistics for subsets of data. See vignette('interval-stats') for
examples.
See also
https://bedtools.readthedocs.io/en/latest/content/tools/jaccard.html
Other interval statistics:
bed_absdist(),
bed_fisher(),
bed_projection(),
bed_reldist()
Examples
genome <- read_genome(valr_example("hg19.chrom.sizes.gz"))
x <- bed_random(genome, seed = 1010486)
y <- bed_random(genome, seed = 9203911)
bed_jaccard(x, y)
#> # A tibble: 1 × 4
#> len_i len_u jaccard n
#> <dbl> <dbl> <dbl> <dbl>
#> 1 236195198 1708776065 0.160 399986
# calculate jaccard per chromosome
bed_jaccard(
dplyr::group_by(x, chrom),
dplyr::group_by(y, chrom)
)
#> # A tibble: 24 × 5
#> chrom len_i len_u jaccard n
#> <chr> <dbl> <dbl> <dbl> <dbl>
#> 1 chr1 18939736 137346996 0.160 32158
#> 2 chr10 10525496 75211235 0.163 17830
#> 3 chr11 10380469 74656102 0.161 17499
#> 4 chr12 10145996 73725426 0.160 17163
#> 5 chr13 8867978 63740684 0.162 14994
#> 6 chr14 8046352 59027220 0.158 13644
#> 7 chr15 7796590 56525195 0.160 13240
#> 8 chr16 6907153 49872823 0.161 11652
#> 9 chr17 6184210 44916521 0.160 10484
#> 10 chr18 6048908 43251434 0.163 10133
#> # ℹ 14 more rows