Question

我想基于字符串列表创建一个二进制矩阵。

dt = data.table(id = c('id1','id2','id3','id4','id5','id6'), sample = c("MER-1,MER-3,MER-4","MER-5","MER-2","MER-2,MER-3,MER-4,MER-5","MER_3","MER-5" ))

dt
    id                  sample
1: id1       MER-1,MER-3,MER-4
2: id2                   MER-5
3: id3                   MER-2
4: id4 MER-2,MER-3,MER-4,MER-5
5: id5                   MER_3
6: id6                   MER-5

应该会导致类似的情况

m_count = matrix(c(1,0,1,1,0, 0,0,0,0,1, 0,1,0,0,0, 0,1,1,1,1, 0,0,1,0,0, 0,0,0,0,1), nrow = 6, ncol = 5)

m_count

    MER-1 MER-2 MER-3 MER-4 MER-5
id1     1     0     0     1     0
id2     0     0     0     1     0
id3     1     0     0     0     0
id4     1     1     0     0     0
id5     0     0     1     1     0
id6     0     1     1     0     1

我可以遍历列表的每个元素，并填充矩阵，但是鉴于表的大小，这确实很慢。有什么更快/更优雅的方法吗？也许与dplyr / tidyverse吗？谢谢！

Answer 1

使用注释末尾的dt来修正问题中的错字，使用separate_rows逐行扩展数据，然后使用table计算计数。< / p>

library(data.table)
library(dplyr)
library(tidyr)

dt %>%
  separate_rows(sample, sep = ",") %>%
  table

给予：

     sample
id    MER-1 MER-2 MER-3 MER-4 MER-5
  id1     1     0     1     1     0
  id2     0     0     0     0     1
  id3     0     1     0     0     0
  id4     0     1     1     1     1
  id5     0     0     1     0     0
  id6     0     0     0     0     1

注意

library(data.table)
dt <- data.table(id = c('id1','id2','id3','id4','id5','id6'), 
  sample = c("MER-1,MER-3,MER-4","MER-5","MER-2","MER-2,MER-3,MER-4,MER-5","MER-3","MER-5" ))

Answer 2

您可以使用strsplit：

table(dt[,unlist(strsplit(sample,",")),by=id])
     V1
id    MER-1 MER-2 MER-3 MER-4 MER-5 MER_3
  id1     1     0     1     1     0     0
  id2     0     0     0     0     1     0
  id3     0     1     0     0     0     0
  id4     0     1     1     1     1     0
  id5     0     0     0     0     0     1
  id6     0     0     0     0     1     0

Answer 3

您还可以使用库splitstackshape：

table(cSplit(dt, "sample", sep = ",", direction = "long"))

     sample
id    MER-1 MER-2 MER-3 MER-4 MER-5
  id1     1     0     1     1     0
  id2     0     0     0     0     1
  id3     0     1     0     0     0
  id4     0     1     1     1     1
  id5     0     0     1     0     0
  id6     0     0     0     0     1

或使用为此场景专门创建的cSplit_e（由@ A5C1D2H2I1M1N2O1R2T1提供）：

cSplit_e(dt, "sample", sep = ",", type = "character", fill = 0, drop = TRUE)

二进制矩阵的字符串列表

3 个答案:

注意