Consider grouping for a column and selecting rows based on other columns in R

Question

My data is dataframe(fpo):

     damIDpoG4 damSirepoG4 damGpoG4 damPhenpoG4 damTBVpoG4 damGBVpoG4
[1,]    450622      430878        4    5.540501   4.260957   3.422568
[2,]    450623      430878        4    3.046358   4.169094   3.528200
[3,]    450625      430878        4    4.515801   4.543196   3.843761
....
[50,]    450626      470878        4    4.798896   4.501067   3.875034
[51,]    450630      470878        4    4.282659   4.388037   3.830042
[52,]    450632      470878        4    3.553223   4.086484   3.571130

I want to select n number (for example 12) from damIDpoG4 for per similar group of damSirepoG4 according to MAX and or 20% damGBVpoG4 . damSirepoG4 contain 250 groups of identical numbers I try:

fpo %>% group_by(fpo[,2]) %>% sample_n(12)

but my answer is not correct. I could not consider max or percent for dplyr
thanks for attention

Answer 1

We need to pass the column name in group_by (assuming that 'fpo' is data.frame/tbl_df and not a matrix )

fpo %>% 
    group_by(damSirepoG4) %>%
    sample_n(12)