tests/testthat/test-BreakpointGRanges.R
In d-cameron/StructuralVariantAnnotation: Variant annotations for structural variants
context('parsing functions')
library(tidyverse)
library(stringr)
example <- readVcf(.testfile("vcf4.2.example.sv.vcf"), "")
simple <- readVcf(.testfile("simple.vcf"), "")
breakend <- readVcf(.testfile("breakend.vcf"), "")
test_that("bedpe2breakpointgr creates unique ids", {
gr <- pairs2breakpointgr(import(.testfile("unnamed.bedpe")))
expect_equal(c("1", "3", "2", "4"), as.character(seqnames(gr)))
expect_equal(c(19356, 1300148+1, 19427, 1302837+1), start(gr))
expect_equal(c(19356, 1300151, 19427, 1302840), end(gr))
expect_equal(c("+", "+", "-", "+"), as.character(strand(gr)))
expect_equal(c(1, 41, 1, 41), gr$QUAL)
expect_equal(c("bedpe1_1", "bedpe2_1", "bedpe1_2", "bedpe2_2"), names(gr))
})
# unpaired breakend
test_that("partner fails if missing mate", {
expect_error(partner(breakpointRanges(breakend)[1,]))
})
requireNamespace("BSgenome.Hsapiens.UCSC.hg19", quietly=FALSE)
hg19 <- BSgenome.Hsapiens.UCSC.hg19::BSgenome.Hsapiens.UCSC.hg19
test_that(".constrict", {
expect_equal(IRanges::width(.constrict(breakpointRanges(breakend))), rep(1, length(breakpointRanges(breakend))))
expect_equal(start(.constrict(breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100100[ . . SVTYPE=BND;CIPOS=0,1;PARID=b",
"chr1 100100 b N ]chr1:100000]N . . SVTYPE=BND;CIPOS=0,1;PARID=a",
"chr1 100000 c N N]chr1:100100] . . SVTYPE=BND;CIPOS=0,1;PARID=d",
"chr1 100100 d N N]chr1:100000] . . SVTYPE=BND;CIPOS=0,1;PARID=c"
))))), c(100000, 100100, # not 100001, 100101
100000, 100101))
gr <- .constrict(breakpointRanges(.testrecord(c(
"chrM 1 a G ]chrM:16571]G . . SVTYPE=BND;PARID=b;CIPOS=-10,-5",
"chrM 16571 b G G[chrM:1[ . . SVTYPE=BND;PARID=a;CIPOS=5,10"
))), hg19)
expect_equal(start(gr), c(1, 16571))
})
test_that("findBreakpointOverlaps", {
expect_equal(as.data.frame(findBreakpointOverlaps(breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]N . . SVTYPE=BND;PARID=a",
"chr1 100000 c N N[chr1:100200[ . . SVTYPE=BND;PARID=d",
"chr1 100200 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c"))),
breakpointRanges(.testrecord(c(
"chr1 100000 c N N[chr1:100200[ . . SVTYPE=BND;PARID=d",
"chr1 100200 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c"))))),
data.frame(queryHits=c(3,4), subjectHits=c(1,2)))
expect_equal(as.data.frame(findBreakpointOverlaps(breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]N . . SVTYPE=BND;PARID=a",
"chr1 100000 c N N[chr1:100200[ . . SVTYPE=BND;PARID=d",
"chr1 100200 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c"))),
breakpointRanges(.testrecord(c(
"chr1 100000 c N N[chr1:100200[ . . SVTYPE=BND;PARID=d",
"chr1 100200 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c"))),
maxgap=2000, sizemargin=NULL)),
data.frame(queryHits=c(1,2,3,4), subjectHits=c(1,2,1,2)))
expect_equal(as.data.frame(findBreakpointOverlaps(breakpointRanges(.testrecord(c(
"chr1 1 a N N[chr1:100100[ . . SVTYPE=BND;CIPOS=0,100000;PARID=b",
"chr1 100100 b N ]chr1:1]N . . SVTYPE=BND;PARID=a"))),
breakpointRanges(.testrecord(c(
"chr1 100000 c N N[chr1:100100[ . . SVTYPE=BND;PARID=d",
"chr1 100100 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c"))))),
data.frame(queryHits=c(1,2), subjectHits=c(1,2)))
expect_equal(as.data.frame(findBreakpointOverlaps(breakpointRanges(.testrecord(c(
"chr1 10000 a N <DEL> . . SVTYPE=DEL;SVLEN=-1000"))),
breakpointRanges(.testrecord(c(
"chr1 10000 c N N[chr1:11000[ . . SVTYPE=BND;PARID=d",
"chr1 11000 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c"))),
maxgap=1)),
data.frame(queryHits=c(1,2), subjectHits=c(1,2)))
expect_equal(as.data.frame(findBreakpointOverlaps(breakpointRanges(.testrecord(c(
"chr1 10000 a N <DEL> . . SVTYPE=DEL;SVLEN=-1000"))),
breakpointRanges(.testrecord(c(
"chr1 11000 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c",
"chr1 10000 c N N[chr1:11000[ . . SVTYPE=BND;PARID=d"))),
maxgap=1)),
data.frame(queryHits=c(1,2), subjectHits=c(2,1)))
})
test_that("findBreakpointOverlaps_match_both_sides", {
gr1 = GRanges(seqnames="1", ranges=IRanges(start=c(1, 100), width=1), strand="+", partner=c("2", "1"))
names(gr1) = c("1", "2")
gr2 = GRanges(seqnames="1", ranges=IRanges(start=c(100, 100), width=1), strand="+", partner=c("2", "1"))
names(gr2) = c("1", "2")
expect_equal(nrow(as.data.frame(findBreakpointOverlaps(gr1, gr2, sizemargin=NULL))), 0)
})
test_that("findBreakpointOverlaps: delly vs truth", {
grdelly <- breakpointRanges(.testrecord(c(
"chr12 6905246 DEL00000292 N <DEL> . PASS PRECISE;CIEND=-52,52;CIPOS=-52,52;SVTYPE=DEL;SVMETHOD=EMBL.DELLYv0.6.8;CHR2=chr12;END=100419669;CT=3to5;INSLEN=0;PE=28;MAPQ=34;SR=30;SRQ=1;CONSENSUS=GTGCATACATTTCAGTGACCCGTTTTAGAAACAGAATTAATATGGTGAATAGAGAAAGAAGAAATCAGTGACTTTGGCCAGGCACAGTAGCTCACATCTGTAATCCCAGCACTTTGGGAGGCTGAGACAGTTGGTTGCTTGAGCCCAGGAGT"
)))
grtruth <- breakpointRanges(.testrecord(c(
"chr12 6905247 truth_8545_o C C[chr12:100419669[ . . EVENT=truth_8545_;MATEID=truth_8545_h;PARID=truth_8545_h;SVTYPE=BND",
"chr12 100419669 truth_8545_h G ]chr12:6905247]g . . EVENT=truth_8545_;MATEID=truth_8545_o;PARID=truth_8545_o;SVTYPE=BND"
)))
hits <- findBreakpointOverlaps(grdelly, grtruth, maxgap=200, ignore.strand=TRUE)
expect_equal(2, nrow(as.data.frame(hits)))
})
test_that("countBreakpointOverlaps", {
expect_equal(countBreakpointOverlaps(breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]N . . SVTYPE=BND;PARID=a",
"chr1 100000 c N N[chr1:100200[ . . SVTYPE=BND;PARID=d",
"chr1 100200 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c"))),
breakpointRanges(.testrecord(c(
"chr1 200000 a N N[chr1:200100[ . . SVTYPE=BND;PARID=b",
"chr1 200100 b N ]chr1:200000]N . . SVTYPE=BND;PARID=a",
"chr1 100000 c N N[chr1:100200[ . . SVTYPE=BND;PARID=d",
"chr1 100200 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c")))),
c(0,0,1,1))
})
test_that("countBreakpointOverlaps uniqueAllocation", {
expect_equal(countBreakpointOverlaps(breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100100[ 1 . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]N 1 . SVTYPE=BND;PARID=a",
"chr1 100000 c N N[chr1:100100[ 2 . SVTYPE=BND;PARID=d",
"chr1 100100 d N ]chr1:100000]N 2 . SVTYPE=BND;PARID=c"))),
breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]N . . SVTYPE=BND;PARID=a",
"chr1 100000 c N N[chr1:100100[ . . SVTYPE=BND;PARID=d",
"chr1 100100 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c"))),
countOnlyBest=TRUE), c(0,0,2,2))
})
test_that("extractBreakpointSequence", {
expect_equal(extractBreakpointSequence(breakpointRanges(.testrecord(c(
# CTC> <TGC
"chr1 100000 a N N[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]N . . SVTYPE=BND;PARID=a"
))), hg19, anchoredBases=3), c("CTCTGC", "GCAGAG"))
expect_equal(extractBreakpointSequence(breakpointRanges(.testrecord(c(
# CTC> AC <TGC
"chr1 100000 a N NAC[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]ACN . . SVTYPE=BND;PARID=a"
))), hg19, anchoredBases=3), c("CTCACTGC", "GCAGTGAG"))
expect_equal(extractBreakpointSequence(breakpointRanges(.testrecord(c(
"chr12 1000000 a N [chr12:2000000[N . . SVTYPE=BND;PARID=b", #GGATA
"chr12 2000000 b N [chr12:1000000[N . . SVTYPE=BND;PARID=a" #GAGAA
))), hg19, anchoredBases=5), c("TATCCGAGAA", "TTCTCGGATA"))
expect_equal(extractBreakpointSequence(breakpointRanges(.testrecord(c(
"chr12 1000000 a N [chr12:2000000[N . . SVTYPE=BND;PARID=b;CIPOS=-115,115",
"chr12 2000000 b N [chr12:1000000[N . . SVTYPE=BND;PARID=a;CIPOS=-115,115"
))), hg19, anchoredBases=5), c("TATCCGAGAA", "TTCTCGGATA"))
expect_equal(extractBreakpointSequence(breakpointRanges(.testrecord(c(
"chr1 9595627 a A A[chr1:9597590[ . . MATEID=b;SVTYPE=BND",
"chr1 9597590 b C ]chr1:9595627]C . . MATEID=a;SVTYPE=BND"
))), hg19, anchoredBases=5)[1], "CTCCAAATCC")
expect_equal(extractBreakpointSequence(breakpointRanges(.testrecord(c(
"chr1 1 a N N[chr1:1[ . . MATEID=b;SVTYPE=BND",
"chr1 1 b N ]chr1:1]N . . MATEID=a;SVTYPE=BND"
))), hg19, anchoredBases=2), c("NNNN", "NNNN"))
expect_equal(extractBreakpointSequence(breakpointRanges(.testrecord(c(
"chr1 1 a N N[chr1:1[ . . MATEID=b;SVTYPE=BND",
"chr1 1 b N ]chr1:1]N . . MATEID=a;SVTYPE=BND"
))), hg19, 0, 0), c("", ""))
expect_equal(extractBreakpointSequence(breakpointRanges(.testrecord(c(
"chr1 1 a N N[chr1:1[ . . MATEID=b;SVTYPE=BND",
"chr1 1 b N ]chr1:1]N . . MATEID=a;SVTYPE=BND"
))), hg19, 0, 1), c("N", "N"))
})
test_that("extractReferenceSequence", {
expect_equal(extractReferenceSequence(breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]N . . SVTYPE=BND;PARID=a"
))), hg19, anchoredBases=3), c("CTCACT", "GCATGG"))
expect_equal(extractReferenceSequence(breakpointRanges(.testrecord(c(
"chr1 100000 a N NAC[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]ACN . . SVTYPE=BND;PARID=a"
))), hg19, anchoredBases=3), c("CTCACT", "GCATGG"))
expect_equal(extractReferenceSequence(breakpointRanges(.testrecord(c(
"chr1 100000 a N NAC[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]ACN . . SVTYPE=BND;PARID=a"
))), hg19, anchoredBases=3, followingBases=5), c("CTCACTAA", "GCATGGCG"))
expect_equal(extractReferenceSequence(breakpointRanges(.testrecord(c(
"chr1 100000 a N NAC[chr1:100100[ . . SVTYPE=BND;PARID=b;CIPOS=-5,5",
"chr1 100100 b N ]chr1:100000]ACN . . SVTYPE=BND;PARID=a;CIPOS=-5,5"
))), hg19, anchoredBases=3, followingBases=5), c("CTCACTAA", "GCATGGCG"))
expect_equal(extractReferenceSequence(breakpointRanges(.testrecord(c(
"chrM 1 a G ]chrM:16571]G . . SVTYPE=BND;PARID=b",
"chrM 16571 b G G[chrM:1[ . . SVTYPE=BND;PARID=a"
))), hg19, anchoredBases=2, followingBases=3), c("TCNNN", "TGNNN"))
})
test_that("calculateReferenceHomology", {
expect_gte(calculateReferenceHomology(breakpointRanges(.testrecord(c(
"chr1 9595527 gridss43448o A A[chr1:9597585[ 1502.22 . CIPOS=-115,115;HOMLEN=230;HOMSEQ=TGGGAGGCTGAGGCAGGCAGATCACTTGAGGCCAGGAGTTCAAGACCAGCCTGGCCAACATGGTGAAACCCTGTCTCTACTAAAAATACAGAAAAATTAGCCAGGCATGGTGGCACGTGCCTGTAATCCAGCTACTCGTGAGGCAGAGGCAGGAGAATTGCTTGAACCCAGGAGGTGGAGGTTGCAGTGAGCTGAGATCATGCCACTGCACTCCAGCCTGGGTGACAGAG;MATEID=gridss43448h;SVTYPE=BND",
"chr1 9597585 gridss43448h C ]chr1:9595527]C 1502.22 . CIPOS=-115,115;HOMLEN=230;HOMSEQ=TGGGAGGCTGAGGCAGGCAGATCACTTGAGGCCAGGAGTTCAAGACCAGCCTGGCCAACATGGTGAAACCCTGTCTCTACTAAAAATACAGAAAAATTAGCCAGGCATGGTGGCACGTGCCTGTAATCCAGCTACTCGTGAGGCAGAGGCAGGAGAATTGCTTGAACCCAGGAGGTGGAGGTTGCAGTGAGCTGAGATCATGCCACTGCACTCCAGCCTGGGTGACAGAG;MATEID=gridss43448o;SVTYPE=BND"
))), hg19)$inexacthomlen[1], 230)
expect_lt(calculateReferenceHomology(breakpointRanges(.testrecord(c(
"chr12 1000000 a A A[chr12:2000000[ . . MATEID=b;SVTYPE=BND",
"chr12 2000000 b C ]chr12:1000000]C . . MATEID=a;SVTYPE=BND"
))), hg19)$inexacthomlen[1], 3)
expect_equal(calculateReferenceHomology(breakpointRanges(.testrecord(c(
"chr1 1 a A A[chr1:1[ . . MATEID=b;SVTYPE=BND",
"chr1 1 b C ]chr1:1]C . . MATEID=a;SVTYPE=BND"
))), hg19, 5)$inexacthomlen, c(NA,NA))
expect_true(is.na(calculateReferenceHomology(breakpointRanges(.testrecord(c(
"chr1 1 a A A[chr1:1[ . . MATEID=b;SVTYPE=BND",
"chr1 1 b C ]chr1:1]C . . MATEID=a;SVTYPE=BND",
"chr12 1000000 aa A A[chr12:2000000[ . . MATEID=bb;SVTYPE=BND",
"chr12 2000000 bb C ]chr12:1000000]C . . MATEID=aa;SVTYPE=BND"
))), hg19, 5)$inexacthomlen[1]))
})
test_that("pairs_round_trip", {
for (f in c("gridss.bedpe", "unnamed.bedpe")) {
pairs = import(system.file("extdata", f, package = "StructuralVariantAnnotation"))
gr = pairs2breakpointgr(pairs)
pairs2 = breakpointgr2pairs(gr)
expect_equal(seqnames(S4Vectors::first(pairs)), seqnames(S4Vectors::first(pairs2)))
expect_equal(start(S4Vectors::first(pairs)), start(S4Vectors::first(pairs2)))
expect_equal(strand(S4Vectors::first(pairs)), strand(S4Vectors::first(pairs2)))
expect_equal(seqnames(S4Vectors::second(pairs)), seqnames(S4Vectors::second(pairs2)))
expect_equal(start(S4Vectors::second(pairs)), start(S4Vectors::second(pairs2)))
expect_equal(strand(S4Vectors::second(pairs)), strand(S4Vectors::second(pairs2)))
expect_equal(mcols(pairs)$name, mcols(pairs2)$name)
expect_equal(mcols(pairs)$score, mcols(pairs2)$score)
}
})
test_that("single_breakends_valididity", {
colo829_vcf = VariantAnnotation::readVcf(system.file("extdata", "COLO829T.purple.sv.ann.vcf.gz", package = "StructuralVariantAnnotation"))
colo829_bpgr <- breakpointRanges(colo829_vcf)
colo829_begr <- breakendRanges(colo829_vcf)
colo829_gr <- sort(c(colo829_begr, colo829_bpgr))
.assertValidBreakpointGRanges(colo829_gr, allowSingleBreakends=TRUE)
expect_error(.assertValidBreakpointGRanges(colo829_gr, allowSingleBreakends=FALSE))
})
#test_that("calculateBlastHomology", {
# Sys.setenv(PATH=paste(Sys.getenv("PATH"), "/usr/local/bioinf/bin", sep=":"))
# bh <- calculateBlastHomology(gr, hg19, "~/blastdb/16SMicrobial")
#
#})
# test_that("performance_test_partner", {
# n = 10000
# gr = GRanges(
# seqnames="1",
# ranges=IRanges(start=1:(2*n), width=1),
# partner=c(paste0(1:n, "o"), paste0(1:n, "h")))
# names(gr)=c(paste0(1:n, "h"), paste0(1:n, "o"))
# tictoc::tic(paste0("Start", n))
# pgr = partner(gr)
# tictoc::toc()
# })
test_that("simpleEventLength", {
expect_equal(simpleEventLength(breakpointRanges(.testrecord(c(
"chr1 100000 a N ]chr1:100009]N . . SVTYPE=BND;PARID=b",
"chr1 100009 b N N[chr1:100000[ . . SVTYPE=BND;PARID=a",
"chr1 100000 c N NNNNNNNNNNN[chr1:100001[ . . SVTYPE=BND;PARID=d",
"chr1 100001 d N ]chr1:100000]NNNNNNNNNNN . . SVTYPE=BND;PARID=c")))),
c(10, 10, 10, 10))
expect_equal(simpleEventLength(breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100002[ . . SVTYPE=BND;PARID=b",
"chr1 100002 b N ]chr1:100000]N . . SVTYPE=BND;PARID=a")))),
c(-1, -1))
})
test_that("simpleEventType", {
expect_equal(simpleEventType(breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]N . . SVTYPE=BND;PARID=a"
)))), c("DEL", "DEL"))
expect_equal(simpleEventType(breakpointRanges(.testrecord(c(
"chr1 100000 a N ]chr1:100100]N . . SVTYPE=BND;PARID=b",
"chr1 100100 b N N[chr1:100000[ . . SVTYPE=BND;PARID=a"
)))), c("DUP", "DUP"))
expect_equal(simpleEventType(breakpointRanges(.testrecord(c(
"chr1 100000 a N NNNNNNNNNNNNNNNNNNNNNNNN[chr1:100001[ . . SVTYPE=BND;PARID=b",
"chr1 100001 b N ]chr1:100000]NNNNNNNNNNNNNNNNNNNNNNNN . . SVTYPE=BND;PARID=a"
)))), c("INS", "INS"))
expect_equal(simpleEventType(breakpointRanges(.testrecord(c(
"chr1 100000 a N NNNNNNNNNNNNNNNNNNNNNNNN[chr2:100001[ . . SVTYPE=BND;PARID=b",
"chr2 100001 b N ]chr1:100000]NNNNNNNNNNNNNNNNNNNNNNNN . . SVTYPE=BND;PARID=a"
)))), c("CTX", "CTX"))
expect_equal(simpleEventType(breakendRanges(.testrecord(c(
"chr1 100000 a N NNNNN. . . SVTYPE=BND"
)))), c("BND"))
})
test_that("findInsDupOverlaps", {
gr1 = breakpointRanges(.testrecord(c(
"chr1 100000 a N ]chr1:100009]N . . SVTYPE=BND;PARID=b",
"chr1 100009 b N N[chr1:100000[ . . SVTYPE=BND;PARID=a",
"chr1 100000 c N NNNNNNNNNNN[chr1:100001[ . . SVTYPE=BND;PARID=d",
"chr1 100001 d N ]chr1:100000]NNNNNNNNNNN . . SVTYPE=BND;PARID=c")))
gr2 = breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]N . . SVTYPE=BND;PARID=a",
"chr1 100000 c N NNNNNNNNNNN[chr1:100001[ . . SVTYPE=BND;PARID=d",
"chr1 100001 d N ]chr1:100000]NNNNNNNNNNN . . SVTYPE=BND;PARID=c")))
hits12 = findInsDupOverlaps(gr1, gr2, maxgap=1)
hits21 = findInsDupOverlaps(gr2, gr1, maxgap=1)
expect_equal(queryHits(hits12), c(1,2))
expect_equal(subjectHits(hits12), c(3,4))
# symmetrical
expect_equal(subjectHits(hits12), queryHits(hits21))
expect_equal(subjectHits(hits21), queryHits(hits12))
})
if (FALSE) {
transitiveGr = bpgr
subjectGr = bpgr
maximumInsertSize=700
maximumTransitiveBreakpoints=4
positionalMargin=8
impreciseTransitiveCalls=(transitiveGr$HOMLEN == 0 | is.null(transitiveGr$HOMLEN)) & start(transitiveGr) != end(transitiveGr)
impreciseSubjectCalls=(subjectGr$HOMLEN == 0 | is.null(subjectGr$HOMLEN)) & start(subjectGr) != end(subjectGr)
allowImprecise=FALSE
}
test_that("findTransitiveCalls imprecise", {
bpgr = breakpointRanges(.testrecord(c(
"chr1 100 bp1_1 N N[chr2:200[ . . SVTYPE=BND;MATEID=bp1_2",
"chr2 200 bp1_2 N ]chr1:100]N . . SVTYPE=BND;MATEID=bp1_1",
"chr2 300 bp2_1 N N[chr3:500[ . . SVTYPE=BND;MATEID=bp2_2",
"chr3 500 bp2_2 N ]chr2:300]N . . SVTYPE=BND;MATEID=bp2_1",
"chr1 50 imprecise_transitive_1 N N[chr3:530[ . . SVTYPE=BND;MATEID=imprecise_transitive_2;IMPRECISE;CIPOS=-49,100",
"chr3 530 imprecise_transitive_2 N ]chr1:50]N . . SVTYPE=BND;MATEID=imprecise_transitive_1;IMPRECISE;CIPOS=-100,100"
)))
transdf = findTransitiveCalls(bpgr, bpgr)
resultdf = DataFrame(
transitive_breakpoint_name=c("imprecise_transitive_1", "imprecise_transitive_2"),
total_distance=101,
traversed_breakpoint_names=CharacterList(c("bp1_1", "bp2_1"), c("bp2_2", "bp1_2")),
distance_to_traversed_breakpoint=IntegerList(c(0, 101), c(0, 101)))
expect_equal(transdf, resultdf)
})
test_that("findTransitiveCalls loop", {
bpgr = breakpointRanges(.testrecord(c(
"chr1 10000 bp1_1 N N[chr2:100[ . . SVTYPE=BND;MATEID=bp1_2",
"chr2 100 bp1_2 N ]chr1:10000]N . . SVTYPE=BND;MATEID=bp1_1",
"chr2 300 bp2_1 N N[chr3:50000[ . . SVTYPE=BND;MATEID=bp2_2",
"chr3 50000 bp2_2 N ]chr2:300]N . . SVTYPE=BND;MATEID=bp2_1",
"chr2 140 loop_1 N ]chr2:160]TTTTTN . . SVTYPE=BND;MATEID=loop_2",
"chr2 160 loop_2 N NTTTTT[chr2:140[ . . SVTYPE=BND;MATEID=loop_1",
"chr1 10050 imprecise_transitive_1 N N[chr3:50030[ . . SVTYPE=BND;MATEID=imprecise_transitive_2;IMPRECISE;CIPOS=-60,60",
"chr3 50030 imprecise_transitive_2 N ]chr1:10050]N . . SVTYPE=BND;MATEID=imprecise_transitive_1;IMPRECISE;CIPOS=-100,100"
)))
.us = function(df) as.data.frame(df) %>% mutate(
traversed_breakpoint_names=unstrsplit(traversed_breakpoint_names, ","),
distance_to_traversed_breakpoint=unstrsplit(CharacterList(distance_to_traversed_breakpoint), ","))
transdf = findTransitiveCalls(bpgr, bpgr, maximumTraversedBreakpoints=5) %>%
.us() %>%
dplyr::arrange(transitive_breakpoint_name, total_distance)
resultdf = DataFrame(
transitive_breakpoint_name=rep(c("imprecise_transitive_1", "imprecise_transitive_2"), each=4),
total_distance=rep(201 + 26*0:3, 2),
traversed_breakpoint_names=CharacterList(
c("bp1_1", "bp2_1"),
c("bp1_1", "loop_2", "bp2_1"),
c("bp1_1", "loop_2","loop_2", "bp2_1"),
c("bp1_1", "loop_2","loop_2","loop_2", "bp2_1"),
c("bp2_2", "bp1_2"),
c("bp2_2", "loop_1", "bp1_2"),
c("bp2_2", "loop_1", "loop_1", "bp1_2"),
c("bp2_2", "loop_1", "loop_1", "loop_1", "bp1_2")),
distance_to_traversed_breakpoint=IntegerList(
cumsum(c(0, 201)),
cumsum(c(0, 61+5, 161)),
cumsum(c(0, 61+5, 21+5, 161)),
cumsum(c(0, 61+5, 21+5, 21+5, 161)),
cumsum(c(0, 201)),
cumsum(c(0, 161+5, 61)),
cumsum(c(0, 161+5, 21+5, 61)),
cumsum(c(0, 161+5, 21+5, 21+5, 61)))) %>%
.us()
expect_equal(transdf, resultdf)
})
test_that(".traversable_segments", {
bpgr = breakpointRanges(.testrecord(c(
"chr1 10000 bp1_1 N N[chr2:100[ . . SVTYPE=BND;MATEID=bp1_2",
"chr2 100 bp1_2 N ]chr1:10000]N . . SVTYPE=BND;MATEID=bp1_1",
"chr2 300 bp2_1 N N[chr3:50000[ . . SVTYPE=BND;MATEID=bp2_2",
"chr3 50000 bp2_2 N ]chr2:300]N . . SVTYPE=BND;MATEID=bp2_1",
"chr2 140 loop_1 N ]chr2:160]TTTTTN . . SVTYPE=BND;MATEID=loop_2",
"chr2 160 loop_2 N NTTTTT[chr2:140[ . . SVTYPE=BND;MATEID=loop_1"
)))
# -1 bp1 -loop-
# | | | +
# 2--------5----6--------3
# - - + |
# 4-- bp2
# 100 140 160 300
# segments are:
# 2----------------------3
# 2-------------6
# 5-------------3
#
bpgr$ordinal = seq_len(length(bpgr))
bpgr$partnerOrdinal = partner(bpgr)$ordinal
segdf = .traversable_segments(bpgr, 1000)
expecteddf = data.frame(
segmentStartExternalOrdinal=c(1, 1, 6, 6),
segmentStartInternalOrdinal=c(2, 2, 5, 5),
segmentStartAdditionalLength=c(0,0,5,5),
segmentLength=c(61, 201, 161, 21),
segmentEndAdditionalLength=c(5, 0, 0, 5),
segmentEndInternalOrdinal=c(6, 3, 3, 6),
segmentEndExternalOrdinal=c(5, 4, 4, 5))
expecteddf = dplyr::bind_rows(expecteddf, expecteddf %>%
dplyr::select(
segmentStartExternalOrdinal=segmentEndExternalOrdinal,
segmentEndExternalOrdinal=segmentStartExternalOrdinal,
segmentLength=segmentLength,
segmentStartInternalOrdinal=segmentEndInternalOrdinal,
segmentEndInternalOrdinal=segmentStartInternalOrdinal,
segmentStartAdditionalLength=segmentEndAdditionalLength,
segmentEndAdditionalLength=segmentStartAdditionalLength))
expect_equal(
segdf %>% dplyr::arrange(segmentStartExternalOrdinal, segmentEndExternalOrdinal, segmentStartInternalOrdinal),
expecteddf %>% dplyr::arrange(segmentStartExternalOrdinal, segmentEndExternalOrdinal, segmentStartInternalOrdinal))
})
test_that("findTransitiveCalls precise", {
hundred_N="NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN"
two_hundred_N=paste0(hundred_N, hundred_N)
# chr1 --]100]
# |
# chr2 [200[--------]300]
# |
# chr3 [100[----]220]
# |
# chr4 [1000[-----
bpgr = breakpointRanges(.testrecord(c(
"chr1 100 bp1_1 N N[chr2:200[ . . SVTYPE=BND;MATEID=bp1_2",
"chr2 200 bp1_2 N ]chr1:100]N . . SVTYPE=BND;MATEID=bp1_1",
"chr2 300 bp2_1 N N[chr3:100[ . . SVTYPE=BND;MATEID=bp2_2",
"chr3 100 bp2_2 N ]chr2:300]N . . SVTYPE=BND;MATEID=bp2_1",
"chr3 220 bp3_1 N N[chr4:1000[ . . SVTYPE=BND;MATEID=bp3_2",
"chr4 1000 bp3_2 N ]chr3:220]N . . SVTYPE=BND;MATEID=bp3_1",
paste0("chr1 100 precise_transitive_1 N N",two_hundred_N,"[chr4:1000[ . . SVTYPE=BND;MATEID=precise_transitive_2"),
paste0("chr4 1000 precise_transitive_2 N ]chr1:100]",two_hundred_N,"N . . SVTYPE=BND;MATEID=precise_transitive_1")
)))
transdf = findTransitiveCalls(bpgr, bpgr)
resultdf = DataFrame(
transitive_breakpoint_name=c("precise_transitive_1", "precise_transitive_2"),
total_distance=101+121,
traversed_breakpoint_names=CharacterList(c("bp1_1", "bp2_1", "bp3_1"), c("bp3_2", "bp2_2", "bp1_2")),
distance_to_traversed_breakpoint=IntegerList(c(0, 101, 101+121), c(0, 121, 121+101)))
expect_equal(transdf, resultdf)
})
test_that("findTransitiveCalls long read colo829", {
t_vcf = list(
truth = readVcf(.testfile("truthset_somaticSVs_COLO829.vcf")),
nanosv = readVcf(.testfile("colo829_nanoSV_truth_overlap.vcf")))
#gridss = readVcf(.testfile("colo829_gridss_truth_overlap.vcf")))
t_bpgr = list()
for (n in names(t_vcf)) {
t_bpgr[[n]] = breakpointRanges(t_vcf[[n]], inferMissingBreakends=TRUE, ignoreUnknownSymbolicAlleles=TRUE)
}
t_bpgr$nanosv$insLen = pmax(0, info(t_vcf$nanosv[t_bpgr$nanosv$sourceId])$GAP)
transdf = findTransitiveCalls(t_bpgr$nanosv, t_bpgr$truth)
expect_equal(4, nrow(transdf))
#transitive_files = list(
# truth = "C:/dev/colo829/truthset_somaticSVs_COLO829.vcf",
# pbsv = "C:/dev/colo829/COLO829.somatic.pacbio.pbsv.vcf",
# nanosv = "C:/dev/colo829/COLO829.nanopore.nanosv.vcf",
# sniffles = "C:/dev/colo829/COLO829.nanopore.sniffles.vcf")
#gridss = "C:/dev/colo829/COLO829v001-colov1_gridss_COLO829v001T.gridss.unfiltered.vcf.gz")
})
d-cameron/StructuralVariantAnnotation documentation built on Dec. 26, 2021, 6:14 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
context('parsing functions')
library(tidyverse)
library(stringr)
example <- readVcf(.testfile("vcf4.2.example.sv.vcf"), "")
simple <- readVcf(.testfile("simple.vcf"), "")
breakend <- readVcf(.testfile("breakend.vcf"), "")
test_that("bedpe2breakpointgr creates unique ids", {
gr <- pairs2breakpointgr(import(.testfile("unnamed.bedpe")))
expect_equal(c("1", "3", "2", "4"), as.character(seqnames(gr)))
expect_equal(c(19356, 1300148+1, 19427, 1302837+1), start(gr))
expect_equal(c(19356, 1300151, 19427, 1302840), end(gr))
expect_equal(c("+", "+", "-", "+"), as.character(strand(gr)))
expect_equal(c(1, 41, 1, 41), gr$QUAL)
expect_equal(c("bedpe1_1", "bedpe2_1", "bedpe1_2", "bedpe2_2"), names(gr))
})
# unpaired breakend
test_that("partner fails if missing mate", {
expect_error(partner(breakpointRanges(breakend)[1,]))
})
requireNamespace("BSgenome.Hsapiens.UCSC.hg19", quietly=FALSE)
hg19 <- BSgenome.Hsapiens.UCSC.hg19::BSgenome.Hsapiens.UCSC.hg19
test_that(".constrict", {
expect_equal(IRanges::width(.constrict(breakpointRanges(breakend))), rep(1, length(breakpointRanges(breakend))))
expect_equal(start(.constrict(breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100100[ . . SVTYPE=BND;CIPOS=0,1;PARID=b",
"chr1 100100 b N ]chr1:100000]N . . SVTYPE=BND;CIPOS=0,1;PARID=a",
"chr1 100000 c N N]chr1:100100] . . SVTYPE=BND;CIPOS=0,1;PARID=d",
"chr1 100100 d N N]chr1:100000] . . SVTYPE=BND;CIPOS=0,1;PARID=c"
))))), c(100000, 100100, # not 100001, 100101
100000, 100101))
gr <- .constrict(breakpointRanges(.testrecord(c(
"chrM 1 a G ]chrM:16571]G . . SVTYPE=BND;PARID=b;CIPOS=-10,-5",
"chrM 16571 b G G[chrM:1[ . . SVTYPE=BND;PARID=a;CIPOS=5,10"
))), hg19)
expect_equal(start(gr), c(1, 16571))
})
test_that("findBreakpointOverlaps", {
expect_equal(as.data.frame(findBreakpointOverlaps(breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]N . . SVTYPE=BND;PARID=a",
"chr1 100000 c N N[chr1:100200[ . . SVTYPE=BND;PARID=d",
"chr1 100200 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c"))),
breakpointRanges(.testrecord(c(
"chr1 100000 c N N[chr1:100200[ . . SVTYPE=BND;PARID=d",
"chr1 100200 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c"))))),
data.frame(queryHits=c(3,4), subjectHits=c(1,2)))
expect_equal(as.data.frame(findBreakpointOverlaps(breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]N . . SVTYPE=BND;PARID=a",
"chr1 100000 c N N[chr1:100200[ . . SVTYPE=BND;PARID=d",
"chr1 100200 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c"))),
breakpointRanges(.testrecord(c(
"chr1 100000 c N N[chr1:100200[ . . SVTYPE=BND;PARID=d",
"chr1 100200 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c"))),
maxgap=2000, sizemargin=NULL)),
data.frame(queryHits=c(1,2,3,4), subjectHits=c(1,2,1,2)))
expect_equal(as.data.frame(findBreakpointOverlaps(breakpointRanges(.testrecord(c(
"chr1 1 a N N[chr1:100100[ . . SVTYPE=BND;CIPOS=0,100000;PARID=b",
"chr1 100100 b N ]chr1:1]N . . SVTYPE=BND;PARID=a"))),
breakpointRanges(.testrecord(c(
"chr1 100000 c N N[chr1:100100[ . . SVTYPE=BND;PARID=d",
"chr1 100100 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c"))))),
data.frame(queryHits=c(1,2), subjectHits=c(1,2)))
expect_equal(as.data.frame(findBreakpointOverlaps(breakpointRanges(.testrecord(c(
"chr1 10000 a N <DEL> . . SVTYPE=DEL;SVLEN=-1000"))),
breakpointRanges(.testrecord(c(
"chr1 10000 c N N[chr1:11000[ . . SVTYPE=BND;PARID=d",
"chr1 11000 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c"))),
maxgap=1)),
data.frame(queryHits=c(1,2), subjectHits=c(1,2)))
expect_equal(as.data.frame(findBreakpointOverlaps(breakpointRanges(.testrecord(c(
"chr1 10000 a N <DEL> . . SVTYPE=DEL;SVLEN=-1000"))),
breakpointRanges(.testrecord(c(
"chr1 11000 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c",
"chr1 10000 c N N[chr1:11000[ . . SVTYPE=BND;PARID=d"))),
maxgap=1)),
data.frame(queryHits=c(1,2), subjectHits=c(2,1)))
})
test_that("findBreakpointOverlaps_match_both_sides", {
gr1 = GRanges(seqnames="1", ranges=IRanges(start=c(1, 100), width=1), strand="+", partner=c("2", "1"))
names(gr1) = c("1", "2")
gr2 = GRanges(seqnames="1", ranges=IRanges(start=c(100, 100), width=1), strand="+", partner=c("2", "1"))
names(gr2) = c("1", "2")
expect_equal(nrow(as.data.frame(findBreakpointOverlaps(gr1, gr2, sizemargin=NULL))), 0)
})
test_that("findBreakpointOverlaps: delly vs truth", {
grdelly <- breakpointRanges(.testrecord(c(
"chr12 6905246 DEL00000292 N <DEL> . PASS PRECISE;CIEND=-52,52;CIPOS=-52,52;SVTYPE=DEL;SVMETHOD=EMBL.DELLYv0.6.8;CHR2=chr12;END=100419669;CT=3to5;INSLEN=0;PE=28;MAPQ=34;SR=30;SRQ=1;CONSENSUS=GTGCATACATTTCAGTGACCCGTTTTAGAAACAGAATTAATATGGTGAATAGAGAAAGAAGAAATCAGTGACTTTGGCCAGGCACAGTAGCTCACATCTGTAATCCCAGCACTTTGGGAGGCTGAGACAGTTGGTTGCTTGAGCCCAGGAGT"
)))
grtruth <- breakpointRanges(.testrecord(c(
"chr12 6905247 truth_8545_o C C[chr12:100419669[ . . EVENT=truth_8545_;MATEID=truth_8545_h;PARID=truth_8545_h;SVTYPE=BND",
"chr12 100419669 truth_8545_h G ]chr12:6905247]g . . EVENT=truth_8545_;MATEID=truth_8545_o;PARID=truth_8545_o;SVTYPE=BND"
)))
hits <- findBreakpointOverlaps(grdelly, grtruth, maxgap=200, ignore.strand=TRUE)
expect_equal(2, nrow(as.data.frame(hits)))
})
test_that("countBreakpointOverlaps", {
expect_equal(countBreakpointOverlaps(breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]N . . SVTYPE=BND;PARID=a",
"chr1 100000 c N N[chr1:100200[ . . SVTYPE=BND;PARID=d",
"chr1 100200 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c"))),
breakpointRanges(.testrecord(c(
"chr1 200000 a N N[chr1:200100[ . . SVTYPE=BND;PARID=b",
"chr1 200100 b N ]chr1:200000]N . . SVTYPE=BND;PARID=a",
"chr1 100000 c N N[chr1:100200[ . . SVTYPE=BND;PARID=d",
"chr1 100200 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c")))),
c(0,0,1,1))
})
test_that("countBreakpointOverlaps uniqueAllocation", {
expect_equal(countBreakpointOverlaps(breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100100[ 1 . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]N 1 . SVTYPE=BND;PARID=a",
"chr1 100000 c N N[chr1:100100[ 2 . SVTYPE=BND;PARID=d",
"chr1 100100 d N ]chr1:100000]N 2 . SVTYPE=BND;PARID=c"))),
breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]N . . SVTYPE=BND;PARID=a",
"chr1 100000 c N N[chr1:100100[ . . SVTYPE=BND;PARID=d",
"chr1 100100 d N ]chr1:100000]N . . SVTYPE=BND;PARID=c"))),
countOnlyBest=TRUE), c(0,0,2,2))
})
test_that("extractBreakpointSequence", {
expect_equal(extractBreakpointSequence(breakpointRanges(.testrecord(c(
# CTC> <TGC
"chr1 100000 a N N[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]N . . SVTYPE=BND;PARID=a"
))), hg19, anchoredBases=3), c("CTCTGC", "GCAGAG"))
expect_equal(extractBreakpointSequence(breakpointRanges(.testrecord(c(
# CTC> AC <TGC
"chr1 100000 a N NAC[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]ACN . . SVTYPE=BND;PARID=a"
))), hg19, anchoredBases=3), c("CTCACTGC", "GCAGTGAG"))
expect_equal(extractBreakpointSequence(breakpointRanges(.testrecord(c(
"chr12 1000000 a N [chr12:2000000[N . . SVTYPE=BND;PARID=b", #GGATA
"chr12 2000000 b N [chr12:1000000[N . . SVTYPE=BND;PARID=a" #GAGAA
))), hg19, anchoredBases=5), c("TATCCGAGAA", "TTCTCGGATA"))
expect_equal(extractBreakpointSequence(breakpointRanges(.testrecord(c(
"chr12 1000000 a N [chr12:2000000[N . . SVTYPE=BND;PARID=b;CIPOS=-115,115",
"chr12 2000000 b N [chr12:1000000[N . . SVTYPE=BND;PARID=a;CIPOS=-115,115"
))), hg19, anchoredBases=5), c("TATCCGAGAA", "TTCTCGGATA"))
expect_equal(extractBreakpointSequence(breakpointRanges(.testrecord(c(
"chr1 9595627 a A A[chr1:9597590[ . . MATEID=b;SVTYPE=BND",
"chr1 9597590 b C ]chr1:9595627]C . . MATEID=a;SVTYPE=BND"
))), hg19, anchoredBases=5)[1], "CTCCAAATCC")
expect_equal(extractBreakpointSequence(breakpointRanges(.testrecord(c(
"chr1 1 a N N[chr1:1[ . . MATEID=b;SVTYPE=BND",
"chr1 1 b N ]chr1:1]N . . MATEID=a;SVTYPE=BND"
))), hg19, anchoredBases=2), c("NNNN", "NNNN"))
expect_equal(extractBreakpointSequence(breakpointRanges(.testrecord(c(
"chr1 1 a N N[chr1:1[ . . MATEID=b;SVTYPE=BND",
"chr1 1 b N ]chr1:1]N . . MATEID=a;SVTYPE=BND"
))), hg19, 0, 0), c("", ""))
expect_equal(extractBreakpointSequence(breakpointRanges(.testrecord(c(
"chr1 1 a N N[chr1:1[ . . MATEID=b;SVTYPE=BND",
"chr1 1 b N ]chr1:1]N . . MATEID=a;SVTYPE=BND"
))), hg19, 0, 1), c("N", "N"))
})
test_that("extractReferenceSequence", {
expect_equal(extractReferenceSequence(breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]N . . SVTYPE=BND;PARID=a"
))), hg19, anchoredBases=3), c("CTCACT", "GCATGG"))
expect_equal(extractReferenceSequence(breakpointRanges(.testrecord(c(
"chr1 100000 a N NAC[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]ACN . . SVTYPE=BND;PARID=a"
))), hg19, anchoredBases=3), c("CTCACT", "GCATGG"))
expect_equal(extractReferenceSequence(breakpointRanges(.testrecord(c(
"chr1 100000 a N NAC[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]ACN . . SVTYPE=BND;PARID=a"
))), hg19, anchoredBases=3, followingBases=5), c("CTCACTAA", "GCATGGCG"))
expect_equal(extractReferenceSequence(breakpointRanges(.testrecord(c(
"chr1 100000 a N NAC[chr1:100100[ . . SVTYPE=BND;PARID=b;CIPOS=-5,5",
"chr1 100100 b N ]chr1:100000]ACN . . SVTYPE=BND;PARID=a;CIPOS=-5,5"
))), hg19, anchoredBases=3, followingBases=5), c("CTCACTAA", "GCATGGCG"))
expect_equal(extractReferenceSequence(breakpointRanges(.testrecord(c(
"chrM 1 a G ]chrM:16571]G . . SVTYPE=BND;PARID=b",
"chrM 16571 b G G[chrM:1[ . . SVTYPE=BND;PARID=a"
))), hg19, anchoredBases=2, followingBases=3), c("TCNNN", "TGNNN"))
})
test_that("calculateReferenceHomology", {
expect_gte(calculateReferenceHomology(breakpointRanges(.testrecord(c(
"chr1 9595527 gridss43448o A A[chr1:9597585[ 1502.22 . CIPOS=-115,115;HOMLEN=230;HOMSEQ=TGGGAGGCTGAGGCAGGCAGATCACTTGAGGCCAGGAGTTCAAGACCAGCCTGGCCAACATGGTGAAACCCTGTCTCTACTAAAAATACAGAAAAATTAGCCAGGCATGGTGGCACGTGCCTGTAATCCAGCTACTCGTGAGGCAGAGGCAGGAGAATTGCTTGAACCCAGGAGGTGGAGGTTGCAGTGAGCTGAGATCATGCCACTGCACTCCAGCCTGGGTGACAGAG;MATEID=gridss43448h;SVTYPE=BND",
"chr1 9597585 gridss43448h C ]chr1:9595527]C 1502.22 . CIPOS=-115,115;HOMLEN=230;HOMSEQ=TGGGAGGCTGAGGCAGGCAGATCACTTGAGGCCAGGAGTTCAAGACCAGCCTGGCCAACATGGTGAAACCCTGTCTCTACTAAAAATACAGAAAAATTAGCCAGGCATGGTGGCACGTGCCTGTAATCCAGCTACTCGTGAGGCAGAGGCAGGAGAATTGCTTGAACCCAGGAGGTGGAGGTTGCAGTGAGCTGAGATCATGCCACTGCACTCCAGCCTGGGTGACAGAG;MATEID=gridss43448o;SVTYPE=BND"
))), hg19)$inexacthomlen[1], 230)
expect_lt(calculateReferenceHomology(breakpointRanges(.testrecord(c(
"chr12 1000000 a A A[chr12:2000000[ . . MATEID=b;SVTYPE=BND",
"chr12 2000000 b C ]chr12:1000000]C . . MATEID=a;SVTYPE=BND"
))), hg19)$inexacthomlen[1], 3)
expect_equal(calculateReferenceHomology(breakpointRanges(.testrecord(c(
"chr1 1 a A A[chr1:1[ . . MATEID=b;SVTYPE=BND",
"chr1 1 b C ]chr1:1]C . . MATEID=a;SVTYPE=BND"
))), hg19, 5)$inexacthomlen, c(NA,NA))
expect_true(is.na(calculateReferenceHomology(breakpointRanges(.testrecord(c(
"chr1 1 a A A[chr1:1[ . . MATEID=b;SVTYPE=BND",
"chr1 1 b C ]chr1:1]C . . MATEID=a;SVTYPE=BND",
"chr12 1000000 aa A A[chr12:2000000[ . . MATEID=bb;SVTYPE=BND",
"chr12 2000000 bb C ]chr12:1000000]C . . MATEID=aa;SVTYPE=BND"
))), hg19, 5)$inexacthomlen[1]))
})
test_that("pairs_round_trip", {
for (f in c("gridss.bedpe", "unnamed.bedpe")) {
pairs = import(system.file("extdata", f, package = "StructuralVariantAnnotation"))
gr = pairs2breakpointgr(pairs)
pairs2 = breakpointgr2pairs(gr)
expect_equal(seqnames(S4Vectors::first(pairs)), seqnames(S4Vectors::first(pairs2)))
expect_equal(start(S4Vectors::first(pairs)), start(S4Vectors::first(pairs2)))
expect_equal(strand(S4Vectors::first(pairs)), strand(S4Vectors::first(pairs2)))
expect_equal(seqnames(S4Vectors::second(pairs)), seqnames(S4Vectors::second(pairs2)))
expect_equal(start(S4Vectors::second(pairs)), start(S4Vectors::second(pairs2)))
expect_equal(strand(S4Vectors::second(pairs)), strand(S4Vectors::second(pairs2)))
expect_equal(mcols(pairs)$name, mcols(pairs2)$name)
expect_equal(mcols(pairs)$score, mcols(pairs2)$score)
}
})
test_that("single_breakends_valididity", {
colo829_vcf = VariantAnnotation::readVcf(system.file("extdata", "COLO829T.purple.sv.ann.vcf.gz", package = "StructuralVariantAnnotation"))
colo829_bpgr <- breakpointRanges(colo829_vcf)
colo829_begr <- breakendRanges(colo829_vcf)
colo829_gr <- sort(c(colo829_begr, colo829_bpgr))
.assertValidBreakpointGRanges(colo829_gr, allowSingleBreakends=TRUE)
expect_error(.assertValidBreakpointGRanges(colo829_gr, allowSingleBreakends=FALSE))
})
#test_that("calculateBlastHomology", {
# Sys.setenv(PATH=paste(Sys.getenv("PATH"), "/usr/local/bioinf/bin", sep=":"))
# bh <- calculateBlastHomology(gr, hg19, "~/blastdb/16SMicrobial")
#
#})
# test_that("performance_test_partner", {
# n = 10000
# gr = GRanges(
# seqnames="1",
# ranges=IRanges(start=1:(2*n), width=1),
# partner=c(paste0(1:n, "o"), paste0(1:n, "h")))
# names(gr)=c(paste0(1:n, "h"), paste0(1:n, "o"))
# tictoc::tic(paste0("Start", n))
# pgr = partner(gr)
# tictoc::toc()
# })
test_that("simpleEventLength", {
expect_equal(simpleEventLength(breakpointRanges(.testrecord(c(
"chr1 100000 a N ]chr1:100009]N . . SVTYPE=BND;PARID=b",
"chr1 100009 b N N[chr1:100000[ . . SVTYPE=BND;PARID=a",
"chr1 100000 c N NNNNNNNNNNN[chr1:100001[ . . SVTYPE=BND;PARID=d",
"chr1 100001 d N ]chr1:100000]NNNNNNNNNNN . . SVTYPE=BND;PARID=c")))),
c(10, 10, 10, 10))
expect_equal(simpleEventLength(breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100002[ . . SVTYPE=BND;PARID=b",
"chr1 100002 b N ]chr1:100000]N . . SVTYPE=BND;PARID=a")))),
c(-1, -1))
})
test_that("simpleEventType", {
expect_equal(simpleEventType(breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]N . . SVTYPE=BND;PARID=a"
)))), c("DEL", "DEL"))
expect_equal(simpleEventType(breakpointRanges(.testrecord(c(
"chr1 100000 a N ]chr1:100100]N . . SVTYPE=BND;PARID=b",
"chr1 100100 b N N[chr1:100000[ . . SVTYPE=BND;PARID=a"
)))), c("DUP", "DUP"))
expect_equal(simpleEventType(breakpointRanges(.testrecord(c(
"chr1 100000 a N NNNNNNNNNNNNNNNNNNNNNNNN[chr1:100001[ . . SVTYPE=BND;PARID=b",
"chr1 100001 b N ]chr1:100000]NNNNNNNNNNNNNNNNNNNNNNNN . . SVTYPE=BND;PARID=a"
)))), c("INS", "INS"))
expect_equal(simpleEventType(breakpointRanges(.testrecord(c(
"chr1 100000 a N NNNNNNNNNNNNNNNNNNNNNNNN[chr2:100001[ . . SVTYPE=BND;PARID=b",
"chr2 100001 b N ]chr1:100000]NNNNNNNNNNNNNNNNNNNNNNNN . . SVTYPE=BND;PARID=a"
)))), c("CTX", "CTX"))
expect_equal(simpleEventType(breakendRanges(.testrecord(c(
"chr1 100000 a N NNNNN. . . SVTYPE=BND"
)))), c("BND"))
})
test_that("findInsDupOverlaps", {
gr1 = breakpointRanges(.testrecord(c(
"chr1 100000 a N ]chr1:100009]N . . SVTYPE=BND;PARID=b",
"chr1 100009 b N N[chr1:100000[ . . SVTYPE=BND;PARID=a",
"chr1 100000 c N NNNNNNNNNNN[chr1:100001[ . . SVTYPE=BND;PARID=d",
"chr1 100001 d N ]chr1:100000]NNNNNNNNNNN . . SVTYPE=BND;PARID=c")))
gr2 = breakpointRanges(.testrecord(c(
"chr1 100000 a N N[chr1:100100[ . . SVTYPE=BND;PARID=b",
"chr1 100100 b N ]chr1:100000]N . . SVTYPE=BND;PARID=a",
"chr1 100000 c N NNNNNNNNNNN[chr1:100001[ . . SVTYPE=BND;PARID=d",
"chr1 100001 d N ]chr1:100000]NNNNNNNNNNN . . SVTYPE=BND;PARID=c")))
hits12 = findInsDupOverlaps(gr1, gr2, maxgap=1)
hits21 = findInsDupOverlaps(gr2, gr1, maxgap=1)
expect_equal(queryHits(hits12), c(1,2))
expect_equal(subjectHits(hits12), c(3,4))
# symmetrical
expect_equal(subjectHits(hits12), queryHits(hits21))
expect_equal(subjectHits(hits21), queryHits(hits12))
})
if (FALSE) {
transitiveGr = bpgr
subjectGr = bpgr
maximumInsertSize=700
maximumTransitiveBreakpoints=4
positionalMargin=8
impreciseTransitiveCalls=(transitiveGr$HOMLEN == 0 | is.null(transitiveGr$HOMLEN)) & start(transitiveGr) != end(transitiveGr)
impreciseSubjectCalls=(subjectGr$HOMLEN == 0 | is.null(subjectGr$HOMLEN)) & start(subjectGr) != end(subjectGr)
allowImprecise=FALSE
}
test_that("findTransitiveCalls imprecise", {
bpgr = breakpointRanges(.testrecord(c(
"chr1 100 bp1_1 N N[chr2:200[ . . SVTYPE=BND;MATEID=bp1_2",
"chr2 200 bp1_2 N ]chr1:100]N . . SVTYPE=BND;MATEID=bp1_1",
"chr2 300 bp2_1 N N[chr3:500[ . . SVTYPE=BND;MATEID=bp2_2",
"chr3 500 bp2_2 N ]chr2:300]N . . SVTYPE=BND;MATEID=bp2_1",
"chr1 50 imprecise_transitive_1 N N[chr3:530[ . . SVTYPE=BND;MATEID=imprecise_transitive_2;IMPRECISE;CIPOS=-49,100",
"chr3 530 imprecise_transitive_2 N ]chr1:50]N . . SVTYPE=BND;MATEID=imprecise_transitive_1;IMPRECISE;CIPOS=-100,100"
)))
transdf = findTransitiveCalls(bpgr, bpgr)
resultdf = DataFrame(
transitive_breakpoint_name=c("imprecise_transitive_1", "imprecise_transitive_2"),
total_distance=101,
traversed_breakpoint_names=CharacterList(c("bp1_1", "bp2_1"), c("bp2_2", "bp1_2")),
distance_to_traversed_breakpoint=IntegerList(c(0, 101), c(0, 101)))
expect_equal(transdf, resultdf)
})
test_that("findTransitiveCalls loop", {
bpgr = breakpointRanges(.testrecord(c(
"chr1 10000 bp1_1 N N[chr2:100[ . . SVTYPE=BND;MATEID=bp1_2",
"chr2 100 bp1_2 N ]chr1:10000]N . . SVTYPE=BND;MATEID=bp1_1",
"chr2 300 bp2_1 N N[chr3:50000[ . . SVTYPE=BND;MATEID=bp2_2",
"chr3 50000 bp2_2 N ]chr2:300]N . . SVTYPE=BND;MATEID=bp2_1",
"chr2 140 loop_1 N ]chr2:160]TTTTTN . . SVTYPE=BND;MATEID=loop_2",
"chr2 160 loop_2 N NTTTTT[chr2:140[ . . SVTYPE=BND;MATEID=loop_1",
"chr1 10050 imprecise_transitive_1 N N[chr3:50030[ . . SVTYPE=BND;MATEID=imprecise_transitive_2;IMPRECISE;CIPOS=-60,60",
"chr3 50030 imprecise_transitive_2 N ]chr1:10050]N . . SVTYPE=BND;MATEID=imprecise_transitive_1;IMPRECISE;CIPOS=-100,100"
)))
.us = function(df) as.data.frame(df) %>% mutate(
traversed_breakpoint_names=unstrsplit(traversed_breakpoint_names, ","),
distance_to_traversed_breakpoint=unstrsplit(CharacterList(distance_to_traversed_breakpoint), ","))
transdf = findTransitiveCalls(bpgr, bpgr, maximumTraversedBreakpoints=5) %>%
.us() %>%
dplyr::arrange(transitive_breakpoint_name, total_distance)
resultdf = DataFrame(
transitive_breakpoint_name=rep(c("imprecise_transitive_1", "imprecise_transitive_2"), each=4),
total_distance=rep(201 + 26*0:3, 2),
traversed_breakpoint_names=CharacterList(
c("bp1_1", "bp2_1"),
c("bp1_1", "loop_2", "bp2_1"),
c("bp1_1", "loop_2","loop_2", "bp2_1"),
c("bp1_1", "loop_2","loop_2","loop_2", "bp2_1"),
c("bp2_2", "bp1_2"),
c("bp2_2", "loop_1", "bp1_2"),
c("bp2_2", "loop_1", "loop_1", "bp1_2"),
c("bp2_2", "loop_1", "loop_1", "loop_1", "bp1_2")),
distance_to_traversed_breakpoint=IntegerList(
cumsum(c(0, 201)),
cumsum(c(0, 61+5, 161)),
cumsum(c(0, 61+5, 21+5, 161)),
cumsum(c(0, 61+5, 21+5, 21+5, 161)),
cumsum(c(0, 201)),
cumsum(c(0, 161+5, 61)),
cumsum(c(0, 161+5, 21+5, 61)),
cumsum(c(0, 161+5, 21+5, 21+5, 61)))) %>%
.us()
expect_equal(transdf, resultdf)
})
test_that(".traversable_segments", {
bpgr = breakpointRanges(.testrecord(c(
"chr1 10000 bp1_1 N N[chr2:100[ . . SVTYPE=BND;MATEID=bp1_2",
"chr2 100 bp1_2 N ]chr1:10000]N . . SVTYPE=BND;MATEID=bp1_1",
"chr2 300 bp2_1 N N[chr3:50000[ . . SVTYPE=BND;MATEID=bp2_2",
"chr3 50000 bp2_2 N ]chr2:300]N . . SVTYPE=BND;MATEID=bp2_1",
"chr2 140 loop_1 N ]chr2:160]TTTTTN . . SVTYPE=BND;MATEID=loop_2",
"chr2 160 loop_2 N NTTTTT[chr2:140[ . . SVTYPE=BND;MATEID=loop_1"
)))
# -1 bp1 -loop-
# | | | +
# 2--------5----6--------3
# - - + |
# 4-- bp2
# 100 140 160 300
# segments are:
# 2----------------------3
# 2-------------6
# 5-------------3
#
bpgr$ordinal = seq_len(length(bpgr))
bpgr$partnerOrdinal = partner(bpgr)$ordinal
segdf = .traversable_segments(bpgr, 1000)
expecteddf = data.frame(
segmentStartExternalOrdinal=c(1, 1, 6, 6),
segmentStartInternalOrdinal=c(2, 2, 5, 5),
segmentStartAdditionalLength=c(0,0,5,5),
segmentLength=c(61, 201, 161, 21),
segmentEndAdditionalLength=c(5, 0, 0, 5),
segmentEndInternalOrdinal=c(6, 3, 3, 6),
segmentEndExternalOrdinal=c(5, 4, 4, 5))
expecteddf = dplyr::bind_rows(expecteddf, expecteddf %>%
dplyr::select(
segmentStartExternalOrdinal=segmentEndExternalOrdinal,
segmentEndExternalOrdinal=segmentStartExternalOrdinal,
segmentLength=segmentLength,
segmentStartInternalOrdinal=segmentEndInternalOrdinal,
segmentEndInternalOrdinal=segmentStartInternalOrdinal,
segmentStartAdditionalLength=segmentEndAdditionalLength,
segmentEndAdditionalLength=segmentStartAdditionalLength))
expect_equal(
segdf %>% dplyr::arrange(segmentStartExternalOrdinal, segmentEndExternalOrdinal, segmentStartInternalOrdinal),
expecteddf %>% dplyr::arrange(segmentStartExternalOrdinal, segmentEndExternalOrdinal, segmentStartInternalOrdinal))
})
test_that("findTransitiveCalls precise", {
hundred_N="NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN"
two_hundred_N=paste0(hundred_N, hundred_N)
# chr1 --]100]
# |
# chr2 [200[--------]300]
# |
# chr3 [100[----]220]
# |
# chr4 [1000[-----
bpgr = breakpointRanges(.testrecord(c(
"chr1 100 bp1_1 N N[chr2:200[ . . SVTYPE=BND;MATEID=bp1_2",
"chr2 200 bp1_2 N ]chr1:100]N . . SVTYPE=BND;MATEID=bp1_1",
"chr2 300 bp2_1 N N[chr3:100[ . . SVTYPE=BND;MATEID=bp2_2",
"chr3 100 bp2_2 N ]chr2:300]N . . SVTYPE=BND;MATEID=bp2_1",
"chr3 220 bp3_1 N N[chr4:1000[ . . SVTYPE=BND;MATEID=bp3_2",
"chr4 1000 bp3_2 N ]chr3:220]N . . SVTYPE=BND;MATEID=bp3_1",
paste0("chr1 100 precise_transitive_1 N N",two_hundred_N,"[chr4:1000[ . . SVTYPE=BND;MATEID=precise_transitive_2"),
paste0("chr4 1000 precise_transitive_2 N ]chr1:100]",two_hundred_N,"N . . SVTYPE=BND;MATEID=precise_transitive_1")
)))
transdf = findTransitiveCalls(bpgr, bpgr)
resultdf = DataFrame(
transitive_breakpoint_name=c("precise_transitive_1", "precise_transitive_2"),
total_distance=101+121,
traversed_breakpoint_names=CharacterList(c("bp1_1", "bp2_1", "bp3_1"), c("bp3_2", "bp2_2", "bp1_2")),
distance_to_traversed_breakpoint=IntegerList(c(0, 101, 101+121), c(0, 121, 121+101)))
expect_equal(transdf, resultdf)
})
test_that("findTransitiveCalls long read colo829", {
t_vcf = list(
truth = readVcf(.testfile("truthset_somaticSVs_COLO829.vcf")),
nanosv = readVcf(.testfile("colo829_nanoSV_truth_overlap.vcf")))
#gridss = readVcf(.testfile("colo829_gridss_truth_overlap.vcf")))
t_bpgr = list()
for (n in names(t_vcf)) {
t_bpgr[[n]] = breakpointRanges(t_vcf[[n]], inferMissingBreakends=TRUE, ignoreUnknownSymbolicAlleles=TRUE)
}
t_bpgr$nanosv$insLen = pmax(0, info(t_vcf$nanosv[t_bpgr$nanosv$sourceId])$GAP)
transdf = findTransitiveCalls(t_bpgr$nanosv, t_bpgr$truth)
expect_equal(4, nrow(transdf))
#transitive_files = list(
# truth = "C:/dev/colo829/truthset_somaticSVs_COLO829.vcf",
# pbsv = "C:/dev/colo829/COLO829.somatic.pacbio.pbsv.vcf",
# nanosv = "C:/dev/colo829/COLO829.nanopore.nanosv.vcf",
# sniffles = "C:/dev/colo829/COLO829.nanopore.sniffles.vcf")
#gridss = "C:/dev/colo829/COLO829v001-colov1_gridss_COLO829v001T.gridss.unfiltered.vcf.gz")
})
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.