如何使用列表或数据框以及要子集化的值来对多个列进行子集化？

Question

我有列表（元素数量奇数）：

my.list = list(col1 = c("CC", "CT", "TT"), 
     col2 = c("GG", "GT"), 
     col3 = c("CC", "CT"),
     col4 = c("CC", "CG", "GG"), 
     col5 = c("AC", "CC"),
     col6 = "GG")

$col1
[1] "CC" "CT" "TT"

$col2
[1] "GG" "GT"

$col3
[1] "CC" "CT"

$col4
[1] "CC" "CG" "GG"

$col5
[1] "AC" "CC"

$col6
[1] "GG"

可以将其转换为数据框：

mylist.df = plyr::ldply(my.list, rbind)
names(mylist.df) <- c("cols","g1", "g2", "g3")

我想使用

mylist

或

mylist.df

对下面的数据框进行子集化。基本上，我想保留每个值至少有一个元素的每一行

mylist

:

df.to.subset = structure(list(IDs = c("ID1", "ID2", "ID3", "ID4", "ID5", "ID6"), 
               gr = c("gr1", "gr1", "gr1", "gr1", "gr1", "gr1"), 
               var = c(-3.451, -3.469, -3.837, -3.344, -3.904, -3.943), 
               col1 = structure(c(1L, 2L, 3L, 1L, 2L, 2L), levels = c("CC", "CT", "TT"), class = "factor"), 
               col2 = structure(c(1L, 1L, 2L, 3L, 3L, 3L), levels = c("GG", "GT", "TT"), class = "factor"), 
               col3 = structure(c(1L, 2L, 1L, 1L, 1L, 1L), levels = c("CC", "CT"), class = "factor"), 
               col4 = structure(c(2L, 2L, 2L, 2L, 2L, 2L), levels = c("CC", "CG", "GG"), class = "factor"), 
               col5 = structure(c(1L, 2L, 2L, 2L, 2L, 2L), levels = c("AC", "CC"), class = "factor"), 
               col6 = structure(c(1L, 1L, 2L, 1L, 1L, 1L), levels = c("GG","AA"), class = "factor")), 
          row.names = c(NA, 
                        -6L), class = c("tbl_df", "tbl", "data.frame"))

  IDs   gr      var col1  col2  col3  col4  col5  col6 
  <chr> <chr> <dbl> <fct> <fct> <fct> <fct> <fct> <fct>
1 ID1   gr1   -3.45 CC    GG    CC    CG    AC    GG   
2 ID2   gr1   -3.47 CT    GG    CT    CG    CC    GG   
3 ID3   gr1   -3.84 TT    GT    CC    CG    CC    AA   
4 ID4   gr1   -3.34 CC    TT    CC    CG    CC    GG   
5 ID5   gr1   -3.90 CT    TT    CC    CG    CC    GG   
6 ID6   gr1   -3.94 CT    TT    CC    CG    CC    GG   

（最终结果是）

  IDs   gr      var col1  col2  col3  col4  col5  col6 
  ID1   gr1   -3.45 CC    GG    CC    CG    AC    GG   
  ID2   gr1   -3.47 CT    GG    CT    CG    CC    GG   

此外，我想重新调整

df.to.subset

中的每一列以匹配此数据框中的级别：

factor.levels.cols = structure(list(cols = c("col1", "col2", "col3", "col4", "col5", "col6"), 
               g1 = c("CC", "GG", "CC", "CC", "AA", "AA"), 
               g2 = c("CT", "GT", "CT", "CG", "AC", "AG"), 
               g3 = c("TT", "TT", "TT", "GG", "CC", "GG")), 
          row.names = c(NA, 6L), class = "data.frame")

  cols g1 g2 g3
1 col1 CC CT TT
2 col2 GG GT TT
3 col3 CC CT TT
4 col4 CC CG GG
5 col5 AA AC CC
6 col6 AA AG GG

这里是否强制使用 for 循环，或者有没有办法让它更快？我有超过 1,000,000 个条目需要修改。

Answer 1

我想你可以像下面这样使用

subset

subset(
    df.to.subset,
    rowMeans(list2DF(Map(`%in%`, df.to.subset[names(my.list)], my.list))) == 1
)

这给出了

# A tibble: 2 × 9
  IDs   gr      var col1  col2  col3  col4  col5  col6 
  <chr> <chr> <dbl> <fct> <fct> <fct> <fct> <fct> <fct>
1 ID1   gr1   -3.45 CC    GG    CC    CG    AC    GG
2 ID2   gr1   -3.47 CT    GG    CT    CG    CC    GG