To continue this thread in case somebody faces the same issue (or as a reference to myself who might need it later), I managed to count unique polymorphisms after I transposed them each one in their own column using the following code data PrWant;
set hivfull.pr_polymorph;
array pr_polymorph{24}pr_polymorph1-pr_polymorph24;
array new {24} $20 _temporary_;
do _n_=1 to 24;
new{_n_}=pr_polymorph{_n_};
end;
call sortc(of new{*});
count=(new{1}>'');
do _n_=2 to 24;
if new{_n_} ne new{_n_-1} then
count + 1;
end;
run; The next step would be to evaluate which polymorphisms are the most common regardless of where they are placed in the column, it is likely that polymorphism A in column 23 could be found in a different column number for another patient, which is why i think it would've been more useful to have each polymorphism as its own column name and then rows would check if it is present or not.
... View more
