R: Sorting columns based on partial match of column names with row names

Question

I have a data frame that can be simplified to look like this (included the dput at the end):

    T2_KL_21  A1_LC_11  W3_FA_22  RR_BI_12  PL_EW_12  RT_LC_22  YU_BI_21
FA  1         2         3         4         5         6         7
BI  1         2         3         4         5         6         7
KL  1         2         3         4         5         6         7
EW  1         2         3         4         5         6         7
LC  1         2         3         4         5         6         7

I would like to sort the columns so that they follow the order of the row names (based on partial match). It would then look like this:

    W3_FA_22  RR_BI_12  YU_BI_21  T2_KL_21  PL_EW_12  A1_LC_11  RT_LC_22
FA  3         4         7         1         5         2         6
BI  3         4         7         1         5         2         6
KL  3         4         7         1         5         2         6
EW  3         4         7         1         5         2         6
LC  3         4         7         1         5         2         6

If more than one column name contains the string in the row names, they should be kept side by side, but the order does not matter.

I have already filtered the columns so that they all contain a match in the row names.

Here is the dput of the data frame:

structure(list(T2_KL_21 = c(1L, 1L, 1L, 1L, 1L), A1_LC_11 = c(2L, 
2L, 2L, 2L, 2L), W3_FA_22 = c(3L, 3L, 3L, 3L, 3L), RR_BI_12 = c(4L, 
4L, 4L, 4L, 4L), PL_EW_12 = c(5L, 5L, 5L, 5L, 5L), RT_LC_22 = c(6L, 
6L, 6L, 6L, 6L), YU_BI_21 = c(7L, 7L, 7L, 7L, 7L)), .Names = c("T2_KL_21", 
"A1_LC_11", "W3_FA_22", "RR_BI_12", "PL_EW_12", "RT_LC_22", "YU_BI_21"
), class = "data.frame", row.names = c("FA", "BI", "KL", "EW", 
"LC"))

I have tried using pmatch, grep and match, with no success.

Any advice will be much appreciated! Thanks

Answer 1

We can loop through the rownames and grep to find the index of the column names that match, unlist and use that to arrange the columns

df1[unlist(lapply(gsub("\\d+", "", row.names(df1)), function(x) grep(x, names(df1))))]
#W3_FA_22 RR_BI_12 YU_BI_21 T2_KL_21 PL_EW_12 A1_LC_11 RT_LC_22
#FA        3        4        7        1        5        2        6
#BI        3        4        7        1        5        2        6
#KL        3        4        7        1        5        2        6
#EW        3        4        7        1        5        2        6
#LC        3        4        7        1        5        2        6