similarity matrix of sequence alignment

[huda]

hi,
I'm PhD student , I'm working in sequence alignment field.

I do alignment , and got similarity matrix.
My problem is I do not know how I can deal with diagonal of this matrix .

is it must be 1? or is it must the original values of the comparison the element with itself.

if I placed it as 1's which represent the fully matching , in this case the other cell will be larger than diagonal.

I do not know how I have to deal with it, because the elements of diagonal will affect the clustering method later

thanks in advance

[balaji]

if you have sequences and you want to cluster them, why dont you cluster them directly why there is a need to do alignment then generate similarity matrix and further do clustering in the end.

[huda]

if you have sequences and you want to cluster them, why dont you cluster them directly why there is a need to do alignment then generate similarity matrix and further do clustering in the end.

thank you , I have to do alignment the sequences , I need to know the similarity among sequences by alignment them.

Anyway, I resolved the problem, because this problem since long time.

Many thanks for advice