predicting pKa from a known a.a. seq?

Mark D. Garfinkel garfinkl at
Wed Feb 2 14:56:48 EST 1994

In article <ming-010294225828 at>
ming at (Ding Ming) writes:

>In article < at>,
> (Ing-Nang Wang) wrote:
>> I wonder if there is any way you can predict the pKa value from a known
>> amino acid seqence? [...]

>Well, you can calculate it by summing up the pKa's of all the charge groups
>in the sequence. But in these days, people didn't do this kind of
>calculation anymore. It is a very simple calculation and you can get the
>result in less than a minute if you use GCG.

        If it's a simple calculation, why depend upon a computer
program to do it?

        I think it's useful exercise to calculate protein pI values
by hand (OK, with calculator). This way one will really learn which
amino acid side chains are ionized (and their charges) or ionizable
under various conditions. Such knowledge goes a long way towards
developing a good grasp of many aspects of protein structure, function
& enzyme reaction mechanisms. We do still teach those things, don't we??

	There's another reason too: programs are sometimes buggy.
There is at least one commercial nucleic acid sequence manipulation
program out there, which I will not name, for which one version did
not subtract properly the water-of-formation when it calculated the
molecular weight of a predicted protein. The only way I figured it
out was because the number given by this program *seemed* wrong, and
so I redid the calculation myself.

Mark D. Garfinkel (e-mail: garfinkl at
My views are my own, which is why they're copyright 1994

More information about the Methods mailing list