hi,

what is the role of statistics to tell if there is a significant relationship between the sequences or if what looks good is just a random occurrence?

 

what is the tool of statistics can be used to know that?

 

many thanks in advance

There is no simple statistical measure that determines the significance. Whether any given similarity is purely by chance depends to a lot of factors, including the type of sequence, length, search space, etc. BLAST has such a measure, for instance, however there is no clear cut-off that you could conveniently use.

Especially if you only got two random sequences, you cannot assign a probability, as you do not know the population from these sequences where fished out (and hence, how much of it was just chance). The only you can describe is the distance, in this particular case.

 

Also, this is more a computational/mathematical problem that is absolutely not related to DNA or biochemistry per se.

Edited October 10, 201114 yr by CharonY