Speed testing plyranges joins at various numbers of features 1e3 to 1e6

hg38_seqlens.tsv

test_plyranges_joins.R

lens <- c("1000","10000","100000","1000000")
for (i in lens) {
  system(paste0("export TEST_SEED=1"))
  system(paste0("granges random-bed --sort --num ",i," hg38_seqlens.tsv > random_bed_",i,"_A.bed"))
  system(paste0("export TEST_SEED=2"))
  system(paste0("granges random-bed --sort --num ",i," hg38_seqlens.tsv > random_bed_",i,"_B.bed"))
}

library(plyranges)
granges_a_3 <- read_bed("random_bed_1000_A.bed")
granges_b_3 <- read_bed("random_bed_1000_B.bed")
granges_a_4 <- read_bed("random_bed_10000_A.bed")
granges_b_4 <- read_bed("random_bed_10000_B.bed")
granges_a_5 <- read_bed("random_bed_100000_A.bed")
granges_b_5 <- read_bed("random_bed_100000_B.bed")
granges_a_6 <- read_bed("random_bed_1000000_A.bed")
granges_b_6 <- read_bed("random_bed_1000000_B.bed")


library(microbenchmark)
microbenchmark(
  list=list(
    ply3 = bquote( join_overlap_inner(granges_a_3, granges_b_3) ),
    ply4 = bquote( join_overlap_inner(granges_a_4, granges_b_4) ),
    ply5 = bquote( join_overlap_inner(granges_a_5, granges_b_5) ),
    ply6 = bquote( join_overlap_inner(granges_a_6, granges_b_6) )
  ), times=10)