Question

I have the following data structure (which was actually created by counting the occurences of “Result” with ddply):

Experiment Result Count
         A      1   123
         A      2   30
         A      3   5
         B      1   120
         B      2   20
         B      3   5
         B      4   1
         B      5   1
         C      1   130
         C      2   21
...

I want to create a similar data frame that groups (calculates sum) of all results that are greater than 2.

Expected outcome:

Experiment Result Count
         A      1   123
         A      2    30
         A     >2     5
         B      1   120
         B      2    20
         B     >2     7
         C      1   130
         C      2    21
...

Probably plyr can do this but I am new to R and have no idea how to use a custom condition (i.e. result 1,2,>2) and not just the distinct values of a column.

Note: I do not mind the name of the new bin (i.e., may be != ‘<2’).

Answer 1

data <- data.frame(Experiment = c("a","a","a","b","b","b"), 
                   Result = c(1,2,3,1,4,5), Count = c(1,4,6,5,3,6))
part1 <- subset(data, Result<=2)
part2 <- cbind(ddply(data, .(Experiment), summarise, 
               Count = sum(Count[Result>2])), Result = ">2")
final <- rbind(part1,part2)
final[with(final, order(Experiment, rev(Result))),]
   Experiment Result Count
1           a      1     1
2           a      2     4
41          a     >2     6
4           b      1     5
5           b     >2     9
data
  Experiment Result Count
1          a      1     1
2          a      2     4
3          a      3     6
4          b      1     5
5          b      4     3
6          b      5     6