amino acid frequency

Shaun D. Black SHAUN at JASON.UTHCT.EDU
Fri Sep 2 16:31:10 EST 1994


Debbie Hodges writes on 2 Sep 1994 10:55:03 GMT:

> Subject: amino acid frequency
> 
> Hi,
> 	I have a feeling that somone may have asked a similar question
> to this before, unfortunately I have no memory of any answers to it.
> What I would like is a table of some sort that gives the relative
> frequencies of the different amino acids in proteins to see how any
> one amino acid composition compares with the 'norm'. 
>    I realise that I would need to be very cautious in any conclusions I 
> made from such a comparison (the table would reflect the choice of
> proteins used to create it). However it would still be interesting
> to have a look.
> 	Either tables or references to them would be appreciated.
> Thanks
> Debbie
> 
Hi All,
 
     First, you're quite right that there are no hard and fast conclusions
that can be drawn about amino acid frequencies in proteins generally.
Second, the frequencies you ask about have been calculated by the Dayhoff
group in their "Atlas of Protein Structure" series; you'll find tables
as I recall in the latter chapters of most of these volumes.  As an 
alternative, you could consult Genbank codon usage tables and calculate
the frequencies for yourself for a given species.  Overall, the rarest
residues are His, Cys, & Trp; the most conserved is Gly.  In E. coli,
the frequencies are (Schulz and Schirmer, "Principles of Protein Structure"
p. 2): A 13.0%, C 1.8%, D+N 9.9%, E+Q 10.8%, F 3.3%, G 7.8%, H 0.7%, I 4.4%,
K 7.0%, L 7.8%, M 3.8%, P 4.6%, R 5.3%, S 6.0%, T 4.6%, V 6.0%, W 1.0%, and
Y 2.2%.  But, any given protein will not necessarily conform to these
conclusions.  Consider the His rich, Pro rich, etc. proteins that have been
characterized.  However, various gene superfamilies have been noted as
having 'characteristic' amino acid compositions, and general conclusions
in this arena may be warranted.  Finally, there is a group studying protein
amino acid compositions on the net (at EMBL).  Their composition server is
at hobohm at embl.heidelberg.de ; I believe you can get information by mailing
a HELP message there.  Bye now,  Shaun   (shaun at jason.uthct.edu)  
 




More information about the Methods mailing list