This directory contains data files and scripts that should enable you
to reproduce some of the results in our paper, "Electrostatic
Calculations of Side-chain pKa Values in Myoglobin and Comparison with
NMR Data for Histidines" by Donald Bashford, David A. Case, Claudio
Dalvit, Linda Tennant and Peter E. Wright (1993) Biochemistry vol. 23,
pp. 8045--8056.

To do the electrostatic calculations, you will need the multimead
program from the MEAD suite, and I shall assume here that you have
read the main MEAD README file.  To get results that match the paper,
multimead should be compiled with the EPSAVE_OLDWAY macro defined
(this is the default in the makefile supplied with MEAD).  To complete
the titration calculation you will need some sort of program for
solving the multi-site titration problem given a set of intrinsic pK
values and site-site interactions.  I recommend the Monte-Carlo method
as described by P. Beroza, D. R. Fredkin, M. Y. Okamura, & G. Feher
(1991) Proc. Natl. Acad. Sci. U.S.A. 88, 5804.  This method is
implemented in a Fortran program, mcti, by P. Beroza
(beroza@scripps.edu).  That program is not supplied here.

This directory includes ".pqr", ".st", ".ogm", ".mgm", and ".sites"
files suitable for calculation of the COx-AmberBondi
structure-parameter combination, which is one of the nine
structure-parameter combinations for which calculations are reported
in our paper cited above.  It also contains some scripts relating to
the re-definition of pKint and site-site interaction values in order
to accommodate the calculation of tautomerism within a
two-state-per-site formulation of the multi-site titration problem.

The basic steps required are shown in the Bourne-shell script,
master.sh.  The scripts assume that multimead and mcti are executables
in the current directory.  You will need about 30 Mb of disk space.

First, run_mol_multimead.sh is run which runs multimead with .sites
files corresponding to various tautomers of histidines.  Only the
first two multimead runs take a lot of computer time (about 6 hours
on an HP 9000/735).  The next 11 runs use data stored in the many
.potat files to avoid calculating potentials on grids.  After this
stage, you should have a set of .pkint files with values that match
the information given in the supplemental tables to our paper.  The
delta delta G sub Born and delta delta G sub back values can be found
in the .el.out files.

Next to run is make-globals.pl which is a perl script (perl is
described in "Programming Perl" by Larry Wall and Randal L Schwartz,
O'Reilly & Assoc., Sebastapol, 1990, and is available at most anonymous
ftp sites that carry GNU software) that munges the output of the
various multimead runs into global .pkint and .g files for the sake of
the tautomerism problem.  See above and see the comments in
make-globals.pl.

The next step, runmcti.sh, won't work unless you have the Beroza
program mcti (see above).  The previous script, make-globals.pl,
produces output that mcti can read, except that I have made a minor
modification to my own copy of mcti to allow longer residue names which
incorporate the residue number information.  This step takes about an
hour on the HP 9000/735.

Collect-curves.pl rearranges the output of mcti and de-munges the
protonation populations of histidines so as to produce titration
curves, pkhalf and Hill plot information on all sites and tautomerism
curves for the histidines.

Finally a grep over the curve file collects the pkhalf information
into pkhalf.out.  The numbers here should match the numbers for
COx-AmberBondi given in Tables IV and V of our paper.

Have fun,
Don Bashford
bashford@scripps.edu
