Calculate mean of group without 0

I have data set like this :

          W   X    Y    Z
A_rep1     0  15   10   2   
A_rep2     2  14    4   8
A_rep3    15   0    8  22
A_rep4    18   3    0   9
A_rep5     0  12    0  22    
B_rep1    13   0   12  80 
B_rep2    14   2   10  45
B_rep3    15  10   12  36 
B_rep4    89  18   22  78 
B_rep5     0  22   16   0
C_rep1    15   0   12   0
C_rep2    12   0   17   5
C_rep3    22   4   50  47
C_rep4     0  14    0   0 
C_rep5     0   5    4  12 

And i want to calculate the mean of each group in every column but without the 0 values
for example for W column group A i want to calculate the mean of A_rep2 A_rep3 A_rep4 (because i have 0 values in A_rep1 and A_rep5
so the result i want is new rows that contain the mean of each group like this

             W          X    Y    Z
A_rep1       0         15   10   2   
A_rep2       2         14    4   8
A_rep3      15          0    8  22
A_rep4      18          3    0   9
A_rep5       0         12    0  22    
B_rep1      13          0   12  80 
B_rep2      14          2   10  45
B_rep3      15         10   12  36 
B_rep4      89         18   22  78 
B_rep5       0         22   16   0
C_rep1      15          0   12   0
C_rep2      12          0   17   5
C_rep3      22          4   50  47
C_rep4       0         14    0   0 
C_rep5       0          5    4  12 
Mean_A      11.66  
Mean_B
Mean_C  

Please tell me if you need any other clarification of explanations

Thank you

Using dplyr,

library(tidyverse)

df %>% 
 rownames_to_column('groups') %>% 
 group_by(groups = sub('_.*', '', groups)) %>% 
 summarise(across(everything(), ~mean(.x[.x != 0])))

# A tibble: 3 × 5
  groups     W     X     Y     Z
  <chr>  <dbl> <dbl> <dbl> <dbl>
1 A       11.7 11     7.33  12.6
2 B       32.8 13    14.4   59.8
3 C       16.3  7.67 20.8   21.3

Your output is not quite clear, so here is an attempt

> df$group=unlist(lapply(strsplit(rownames(df),"_"),"[[",1))
> aggregate(.~group,data=df,FUN=function(x){mean(x[x!=0])})

  group        W         X         Y        Z
1     A 11.66667 11.000000  7.333333 12.60000
2     B 32.75000 13.000000 14.400000 59.75000
3     C 16.33333  7.666667 20.750000 21.33333

Here's a dplyr solution. You can use separate_wider_delim() to separate the group letter from the rest of the observation identifier. Then, grouping by the group letter, you can summarize across all variables, taking the mean of only those values of the variable not equal to zero. Next, you can change the group values to append them with "Mean ". Finally, you can append the summarized data to the original data.

library(dplyr)
library(tidyr)
tab <- read.table(header=TRUE, textConnection("
 W   X    Y    Z
A_rep1     0  15   10   2   
A_rep2     2  14    4   8
A_rep3    15   0    8  22
A_rep4    18   3    0   9
A_rep5     0  12    0  22    
B_rep1    13   0   12  80 
B_rep2    14   2   10  45
B_rep3    15  10   12  36 
B_rep4    89  18   22  78 
B_rep5     0  22   16   0
C_rep1    15   0   12   0
C_rep2    12   0   17   5
C_rep3    22   4   50  47
C_rep4     0  14    0   0 
C_rep5     0   5    4  12")) 


tab <- tab %>% 
  as_tibble(rownames = "obs") %>% 
  separate_wider_delim(obs, names = c("group", "rep"), delim="_") 
tab %>% 
  group_by(group) %>% 
  summarise(across(W:Z, function(x)mean(x[which(x != 0)]))) %>% 
  mutate(group = paste("Mean ", group, sep="")) %>% 
  bind_rows(tab, .)
#> # A tibble: 18 × 6
#>    group  rep       W     X     Y     Z
#>    <chr>  <chr> <dbl> <dbl> <dbl> <dbl>
#>  1 A      rep1    0   15    10      2  
#>  2 A      rep2    2   14     4      8  
#>  3 A      rep3   15    0     8     22  
#>  4 A      rep4   18    3     0      9  
#>  5 A      rep5    0   12     0     22  
#>  6 B      rep1   13    0    12     80  
#>  7 B      rep2   14    2    10     45  
#>  8 B      rep3   15   10    12     36  
#>  9 B      rep4   89   18    22     78  
#> 10 B      rep5    0   22    16      0  
#> 11 C      rep1   15    0    12      0  
#> 12 C      rep2   12    0    17      5  
#> 13 C      rep3   22    4    50     47  
#> 14 C      rep4    0   14     0      0  
#> 15 C      rep5    0    5     4     12  
#> 16 Mean A <NA>   11.7 11     7.33  12.6
#> 17 Mean B <NA>   32.8 13    14.4   59.8
#> 18 Mean C <NA>   16.3  7.67 20.8   21.3

^{Created on 2023-06-29 with reprex v2.0.2}