Thank you. But I don't want to split these sites into multiple sites. I want the number of lines remain same as before. What I want to do finally is association test, but the association tool will handle these multiple variants as independent, though they are actually from same position.
So the result file I want to have from the example above is like REF=Del, ALT=Ins and T,TT,TTT will be coded as Del, and TTTT, TTTTT should be Ins.
Thanks.
In my opinion, we can separate these multi-allelic variants and do the association separately with specific statistic method.