Question

我的数据集包含以下部分：

structure(list(domain = c("A1BG_-_-_0", "A1BG_-_-_1", "A1BG_-_-_2", 
"A1BG_-_-_3", "A1BG_-_-_4", "A1BG_143228_143228_0", "A1BG_143228_143228_1", 
"A1BG_143228_143228_2", "A1BG_143228_143228_3", "A1CF_-_-_0"), 
    chr = c("19", "19", "19", "19", "19", "19", "19", "19", "19", 
    "10"), positions = c("(58858387..58858395,58858718..58858719)", 
    "(58858998..58859006,58861735..58862017,58862756..58862766)", 
    "(58863018..58863053,58863648..58863673)", "(58863913..58863921,58864293..58864303)", 
    "(58864552..58864563,58864657..58864693,58864769..58864803)", 
    "(58858719..58858998)", "(58862766..58863018)", "(58863673..58863913)", 
    "(58864303..58864552)", "(52566488..52566640,52569653..52569717)"
    ), length = c(11L, 303L, 62L, 20L, 84L, 280L, 253L, 241L, 
    250L, 218L)), class = "data.frame", row.names = c(NA, -10L
))

positions列指定由逗号分隔的一个或多个start..stop位置的序列。

此外，我还有一个位置数据集（显示了一部分）：

structure(list(VarID = 1:9, chr = c(19L, 19L, 19L, 19L, 19L, 
19L, 19L, 19L, 10L), position = c(58864801, 58863673, 58863673, 58863673, 
58863673, 58863673, 58863673, 58863041, 52569689)), class = "data.frame", row.names = c(NA, 
-9L))

我想在第二个数据集后面附加一列，以指定domain所属的VarID。

我想要的输出是：

structure(list(VarID = 1:9, chr = c(19L, 19L, 19L, 19L, 19L, 
19L, 19L, 19L, 10L), position = c(58864801, 58863673, 58863673, 
58863673, 58863673, 58863673, 58863673, 58863041, 52569689), 
    domain = c("A1BG_-_-_4", "A1BG_-_-_2", "A1BG_-_-_2", "A1BG_-_-_2", 
    "A1BG_-_-_2", "A1BG_-_-_2", "A1BG_-_-_2", "A1BG_-_-_2", "A1CF_-_-_0"
    )), row.names = c(NA, -9L), class = "data.frame")

特别是，我无法使gsub正常工作，这最终使我可以查询职位是否在start..stop范围内。

Answer 1

尝试foverlaps中的data.table：

library(data.table)

dtt.domain <- fread(text = '      domain chr                                                  positions length
            A1BG_-_-_0  19                    (58858387..58858395,58858718..58858719)     11
    A1BG_-_-_1  19 (58858998..58859006,58861735..58862017,58862756..58862766)    303
    A1BG_-_-_2  19                    (58863018..58863053,58863648..58863673)     62
    A1BG_-_-_3  19                    (58863913..58863921,58864293..58864303)     20
    A1BG_-_-_4  19 (58864552..58864563,58864657..58864693,58864769..58864803)     84
    A1BG_143228_143228_0  19                                       (58858719..58858998)    280
    A1BG_143228_143228_1  19                                       (58862766..58863018)    253
    A1BG_143228_143228_2  19                                       (58863673..58863913)    241
    A1BG_143228_143228_3  19                                       (58864303..58864552)    250
    A1CF_-_-_0  10                    (52566488..52566640,52569653..52569717)    218')
dtt.var <- fread(text = 'VarID chr position
      1  19 58864801
    2  19 58863673
    3  19 58863673
    4  19 58863673
    5  19 58863673
    6  19 58863673
    7  19 58863673
    8  19 58863041
    9  10 52569689')

dtt.domain2 <- dtt.domain[, .(
    region = strsplit(
        gsub('\\(|\\)', '', positions), ',', fixed = TRUE)[[1]]),
    by = .(domain, chr)]

dtt.domain2[, c('start', 'end') := tstrsplit(
    region, '..', fixed = TRUE, type.convert = TRUE)]

setkeyv(dtt.domain2, c('chr', 'start', 'end'))

dtt.var[, `:=`(start = position, end = position)]

res <- foverlaps(dtt.var, dtt.domain2, mult = 'first')
res[, .(VarID, chr, position, domain)]
#    VarID chr position     domain
# 1:     1  19 58864801 A1BG_-_-_4
# 2:     2  19 58863673 A1BG_-_-_2
# 3:     3  19 58863673 A1BG_-_-_2
# 4:     4  19 58863673 A1BG_-_-_2
# 5:     5  19 58863673 A1BG_-_-_2
# 6:     6  19 58863673 A1BG_-_-_2
# 7:     7  19 58863673 A1BG_-_-_2
# 8:     8  19 58863041 A1BG_-_-_2
# 9:     9  10 52569689 A1CF_-_-_0

Answer 2

另一个使用dplyr和tidyr的选项

library(dplyr)
library(tidyr)

df %>%
   mutate(positions = gsub("[()]", "", positions), 
          chr = as.integer(chr)) %>%
   separate_rows(positions, sep = ",") %>%
   separate(positions, c("start", "end"), sep = "\\.\\.") %>%
   left_join(locations, by = c("chr" = "chrno")) %>%
   filter(loc > start & loc <= end) %>%
   arrange(VarID) %>%
   dplyr::select(VarID, chr, loc, domain)


#  VarID chr      loc     domain
#1     1  19 58864801 A1BG_-_-_4
#2     2  19 58863673 A1BG_-_-_2
#3     3  19 58863673 A1BG_-_-_2
#4     4  19 58863673 A1BG_-_-_2
#5     5  19 58863673 A1BG_-_-_2
#6     6  19 58863673 A1BG_-_-_2
#7     7  19 58863673 A1BG_-_-_2
#8     8  19 58863041 A1BG_-_-_2
#9     9  10 52569689 A1CF_-_-_0

从字符串中提取数字并检查条件

3 个答案: