我有两个Granges对象，我希望将它们合并，以便在两个对象中都不存在元数据的情况下组合两个GRange。
> t
GRanges object with 2 ranges and 5 metadata columns:
seqnames ranges strand | pvalue qvalue meth.diff gc.X gc.score
<Rle> <IRanges> <Rle> | <numeric> <numeric> <numeric> <GRanges> <numeric>
[1] MT [ 708, 708] + | 2.898639e-04 0.007018699 0.2231039 MT:706-710 80
[2] MT [1147, 1147] - | 6.043240e-05 0.003882324 0.2243177 MT:1146-1150 80
-------
seqinfo: 1 sequence from an unspecified genome; no seqlengths
> s
GRanges object with 1 range and 6 metadata columns:
seqnames ranges strand | pvalue qvalue meth.diff gc.X gc.name gc.score
<Rle> <IRanges> <Rle> | <numeric> <numeric> <numeric> <GRanges> <character> <numeric>
[1] MT [708, 708] + | 0.0002898639 0.007018699 0.2231039 MT:708:+ rs28412942 0
-------
seqinfo: 1 sequence from an unspecified genome; no seqlengths
我想在最后有：
> combined
GRanges object with 2 ranges and 5 metadata columns:
seqnames ranges strand | pvalue qvalue meth.diff gc.X gc.score gc.name
<Rle> <IRanges> <Rle> | <numeric> <numeric> <numeric> <GRanges> <numeric>
[1] MT [ 708, 708] + | 2.898639e-04 0.007018699 0.2231039 MT:706-710 80 rs28412942
[2] MT [1147, 1147] - | 6.043240e-05 0.003882324 0.2243177 MT:1146-1150 80 NA
-------
seqinfo: 1 sequence from an unspecified genome; no seqlengths
我尝试过的所有方法都要求具有相同数量的列。我可以尝试创建所需的列并用NA填充，但在我看来这有点过头了，我确信有一个方法存在，但我找不到它：/
谢谢！
使用findOverlaps很容易做到这一点
library(GenomicRanges)
m <- findOverlaps(s, t)
# Add gc.name to subject GRanges (i.e. left join)
mcols(t)$gc.name <- NA
mcols(t)[subjectHits(m), "gc.name"] = mcols(s)[queryHits(m), "gc.name"]
t
#GRanges object with 2 ranges and 6 metadata columns:
# seqnames ranges strand | pvalue qvalue meth.diff gc.X
# <Rle> <IRanges> <Rle> | <numeric> <numeric> <numeric> <character>
# [1] MT [ 708, 708] + | 0.3361535 0.06058539 0.4743142 a
# [2] MT [1147, 1147] - | 0.4637233 0.19743361 0.3010486 b
# gc.score gc.name
# <numeric> <character>
# [1] 80 rs28412942
# [2] 80 <NA>
# -------
# seqinfo: 1 sequence from an unspecified genome; no seqlengths
样本数据
set.seed(2018)
t <- GRanges(
seqnames = c("MT", "MT"),
IRanges(start = c(708, 1147), end = c(708, 1147)),
strand = c("+", "-"),
pvalue = runif(2),
qvalue = runif(2),
meth.diff = runif(2),
gc.X = letters[1:2],
gc.score = rep(80, 2))
set.seed(2018)
s <- GRanges(
seqnames = "MT",
IRanges(start = 708, end = 708),
strand = "+",
pvalue = runif(1),
qvalue = runif(1),
meth.diff = runif(1),
gc.X = letters[3],
gc.name = "rs28412942",
gc.score = 0)
如果有多个重叠，答案表单@Mautirus-Evers将失败。findOverlaps正确地识别了两个重叠，但由于输入查询只有一行，所以具有mcols的第二部分只指定了重叠中的第二个。
"；失败"；测试：
set.seed(2018)
t <- GRanges(
seqnames = "MT",
IRanges(start = 708, end = 708),
strand = "+",
pvalue = runif(1),
qvalue = runif(1),
meth.diff = runif(1),
gc.X = letters[1],
gc.score = rep(80, 1))
t
# GRanges object with 1 range and 5 metadata columns:
# seqnames ranges strand | pvalue qvalue meth.diff gc.X gc.score
# <Rle> <IRanges> <Rle> | <numeric> <numeric> <numeric> <character> <numeric>
# [1] MT 708 + | 0.336153471143916 0.463723270455375 0.0605853858869523 a 80
# -------
# seqinfo: 1 sequence from an unspecified genome; no seqlengths
set.seed(2018)
s <- GRanges(
seqnames = "MT",
IRanges(start = c(708, 708), end = c(708, 708)),
strand = c("+", "+"),
pvalue = runif(2),
qvalue = runif(2),
meth.diff = runif(2),
gc.X = letters[3:4],
gc.name = c("rs28412942", "rs28412942dupl"),
gc.score = 0)
s
# GRanges object with 2 ranges and 6 metadata columns:
# seqnames ranges strand | pvalue qvalue meth.diff gc.X gc.name gc.score
# <Rle> <IRanges> <Rle> | <numeric> <numeric> <numeric> <character> <character> <numeric>
# [1] MT 708 + | 0.336153471143916 0.0605853858869523 0.474314193474129 c rs28412942 0
# [2] MT 708 + | 0.463723270455375 0.197433612542227 0.301048603840172 d rs28412942dupl 0
# -------
# seqinfo: 1 sequence from an unspecified genome; no seqlengths
library(GenomicRanges)
m <- findOverlaps(s, t, type = "any", select = "all")
m
#Hits object with 2 hits and 0 metadata columns:
# queryHits subjectHits
# <integer> <integer>
# [1] 1 1
# [2] 2 1
# -------
# queryLength: 2 / subjectLength: 1
# Add gc.name to subject GRanges (i.e. left join)
mcols(t)$gc.name <- NA
mcols(t)[subjectHits(m), "gc.name"] <- mcols(s)[queryHits(m), "gc.name"]
t
# GRanges object with 1 range and 6 metadata columns:
# seqnames ranges strand | pvalue qvalue meth.diff gc.X gc.score gc.name
# <Rle> <IRanges> <Rle> | <numeric> <numeric> <numeric> <character> <numeric> <character>
# [1] MT 708 + | 0.336153471143916 0.463723270455375 0.0605853858869523 a 80 rs28412942dupl
# -------
# seqinfo: 1 sequence from an unspecified genome; no seqlengths
我最终使用了plyranges join_overlap_left_directed(或join_overlap_left(，然后过滤我需要保留的任何列。
plyranges解决方案：
set.seed(2018)
t <- GRanges(
seqnames = c("MT", "MT"),
IRanges(start = c(708, 800), end = c(708, 800)),
strand = c("+", "-"),
pvalue = runif(2),
qvalue = runif(2),
meth.diff = runif(2),
gc.X = letters[1:2],
gc.score = rep(80, 2))
set.seed(2018)
s <- GRanges(
seqnames = "MT",
IRanges(start = c(708, 708), end = c(708, 708)),
strand = c("+", "+"),
pvalue = runif(2),
qvalue = runif(2),
meth.diff = runif(2),
gc.X = letters[3:4],
gc.name = c("rs28412942", "rs28412942dupl"),
gc.score = 0)
library(plyranges)
t <- join_overlap_left_directed(t, s)
t
# GRanges object with 3 ranges and 11 metadata columns:
# seqnames ranges strand | pvalue.x qvalue.x meth.diff.x gc.X.x gc.score.x pvalue.y qvalue.y meth.diff.y gc.X.y gc.name gc.score.y
# <Rle> <IRanges> <Rle> | <numeric> <numeric> <numeric> <character> <numeric> <numeric> <numeric> <numeric> <character> <character> <numeric>
# [1] MT 708 + | 0.336153471143916 0.0605853858869523 0.474314193474129 a 80 0.336153471143916 0.0605853858869523 0.474314193474129 c rs28412942 0
# [2] MT 708 + | 0.336153471143916 0.0605853858869523 0.474314193474129 a 80 0.463723270455375 0.197433612542227 0.301048603840172 d rs28412942dupl 0
# [3] MT 800 - | 0.463723270455375 0.197433612542227 0.301048603840172 b 80 <NA> <NA> <NA> <NA> <NA> <NA>
# -------
# seqinfo: 2 sequences from an unspecified genome; no seqlengths
附言：我很抱歉将此添加为答案，但由于我的声誉不佳，我无法发表评论。