Python Programming for Biology: Bioinformatics and Beyond

Similarity measures

The first Python examples in this chapter involve measuring values for the similarity and

hence  conservation  of  sequences  in  a  multiple  alignment.  This  is  achieved  by  looking  at

the various positions of the alignment and how similar the residues are down each column.

This  is  not  necessarily  the  best  way  to  measure  conservation  (and  consequently  the

opposite  measure  of  variability),  because  the  choice  of  sequences  can  be  subjective  and

treating  all  sequences  equally  does  not  account  for  the  expectation  in  the  amount  of

change; the more distantly related sequences are as a whole the more change is expected.

Nevertheless  if  we  choose  sequences  carefully  and  know  that  the  evolutionary  distance

between them is appropriate for the question we wish to ask then this type of analysis is

pertinent, and quicker to perform than many more complex methods.