distinct/grouping

[copper]

I have a set of data and it look like this
group1
c1 c2 c3 c4
1 2 3 4
1 2 3 5
1 2 4 6
1 2 4 7
1 3 5 8 and so on for my report

i want to sum or count each column according to their grouping without counting the dupl in each group. i.e i want to add all of c4, add just the unique number in c3 and count only the unique number in c2 in customer group

thanks copper