Equivalent amino acids for similarity searches

Keith Robison robison1 at husc10.harvard.edu
Tue Mar 22 22:36:03 EST 1994


dy684 at cleveland.Freenet.Edu (Roberto Albinoni) writes:

>i am comparing the dna-derived amino acids sequences of two proteins in
>order to determine which residues may be most important to enzymatic
>function. 

>i wonder if someone to point me to appropriate references on which amino
>acids can be considered most similar based on their chemistry and function,
>and also how one can 'score' the similarity of two proteins based not only
>on identical residues, but also on amino acids which belong to the same
>chemical families.

Producing such "substitution matrices" has been a cottage industry for
some time.  Steve Henikoff has developed what are probably the best
general-purpose matrices (BLOSUM) and has published papers analyzing
the practical differences between many of these; much of the theoretical
background for this work was done by Steven Altschul.  

You should try looking in Amos Bairoch's SeqAnalRef.  It's available
via gopher from gopher.gdb.org, and via World-Wide Web (e.g. Mosaic)
from http://golgi.harvard.edu/gilbert-bi.html.

I have started writing a hypertext guide to sequence searching issues
& terms.  The chapter on substitution matrices is incomplete, but
you will find many starting references in it.  The guide is available
at http://twod.med.harvard.edu/seqanal
(Yes, this URL really works, as opposed to the last time I posted it :-)


Keith Robison
Harvard University
Department of Cellular and Developmental Biology
Department of Genetics / HHMI

krobison at nucleus.harvard.edu 









More information about the Proteins mailing list