ChangeLog.c29
  CHARMM 29
  c29a1   Developmental   August 15, 2001
  c29a2   Developmental   April 15, 2002
  c29b0   Developmental   July 22, 2002
  c29b1   Release   August 15, 2002
  c29b2   Release   February 15, 2003
  About CHARMM29 Development   Next Top

As an on-going project, CHARMM development will be carried out with
CHARMM version 29 series.  The base revision is the beta release version
c28b1.  c29a1 is the first revision in the c29 developmental versions.
The following describes new features and enhancements incorporated in
the c29 development period.

  1. c29a1 Change Log Previous Next Top

    The following are introduced and modified after the c28b1 release
dated August 15, 2001.

1.1 Enhancements to the Monte Carlo and Dynamics modules
    Name          : Aaron R. Dinner, Ioan Andricioaei, and Martin Karplus
    Email Address : aaron@bellatrix.pcl.ox.ac.uk
    Institution   : University of Oxford
    Date          : June 15, 2001

1) Volume scaling moves for constant pressure MC simulations were added.
The method of choosing the volume changes and the acceptance criterion 
are those introduced by Eppenga and Frenkel (see mc.doc).  Scaling can be 
applied BYATom, BYREsidue, or BYHEavy (the last maintains fixed hydrogen 
bond lengths).  These modes of scaling can be mixed with the MOVE LINK 
command described below.  The method works with images generated either by 
the CRYStal command (now officially supported for use with MC) or by a series 
of IMAGE TRANSformations (in which case the initial volume must be specified).

2) Dynamics restart files can now be written and read by MC, largely to make
it easier to extend constant pressure simulations via storage of the CRYStal
data structure. 

3) A general MOVE LINKing facility was introduced to allow specification that 
certain MC moves always be performed together before applying the acceptance 
criterion.  For example, one might wish to perform both a rigid body 
translation and a rigid body rotation of a solvent molecule in the same MC 
step; to do so, one would define each of these types of moves with MOVE ADD
and then tie them to each other with MOVE LINK.  At present, move instances 
within the move groups selected are matched by indices, so the move groups 
must have the same numbers of instances.  Further details of the use of MOVE 
LINK are given in mc.doc.

4) Molecular dynamics simulations and Hybrid Monte Carlo simulations in the 
Tsallis (generalized) ensemble can now be performed.  The modifications
(which require the keyword TSALLIS to be included in pref.dat) involved a 
change to the calling sequence of DYNAMC.  To invoke the Tsallis method in
molecular dynamics, it is simply necessary to add to the DYNAmics command
the keywords:  TSALlis QTSAllis real EMIN real (see dynamc.doc).  The Tsallis
transformation is applied automatically in Hybrid Monte Carlo simulations if
IACCept=2 (see mc.doc).

5) The path integral energy term is now supported by Monte Carlo.  Small
changes were made to the existing path integral setup routines (to create
additional arrays for faster MC calculations, invoked by including the
keyword MC on the PINT commandline) and to the existing path integral energy
routine (to eliminate two unnecessary SQRT operations in the pairwise loop).
In addition, an MC specific path integral energy routine (EPIMC) was
introduced to pathint.src.  At present, the allowed path integral moves are
displacements of single beads (RTRN BYATom) or movements of five sequential
beads such that the path integral spring lengths remain unchanged (CROT PIMC)
(see replica.doc and mc.doc).  
 
6) MC simulations with the semiempirical Mopac QM/MM terms are now supported.


Files Added:
source/mc/moveln.src          (3)

Files Modified:
source/charmm/charmm_main.src (3)
      /charmm/iniall.src      (1, 3)
      /dynamc/dcntrl.src      (4)
      /dynamc/dynamc.src      (4)
      /energy/pathint.src     (5)
      /energy/polar.src       (5)
      /energy/printe.src      (1)
      /misc/pbeq2.src         (5)

      /mc/armupt.src          
      /mc/mc.src              (1, 3, 4, 5)
      /mc/mcener.src          (1, 5, 6)
      /mc/mcio.src            (2)
      /mc/movead.src          (1, 3, 5)
      /mc/moveio.src          (1, 3, 5)
      /mc/mvcrot.src          (5)
      /mc/mvdihe.src          (5) 
      /mc/mvhmc.src           (4, 5)
      /mc/mvrtrn.src          (1, 5)

      /fcm/mc.fcm             (3, 5)
      /fcm/consta.fcm         (5)
      /fcm/pathint.fcm        (5)

Modified Documentation:
doc/dynamc.doc, mc.doc, replica.doc

New Testcase:
test/c29test/mctest03.inp, mctest04.inp, mctest05.inp, mctest06.inp
            /mctest07.inp

------------------------------------------------------------------------------

1.2 The SASA Implicit Solvation Model
    Name          : Urs Haberthuer and Amedeo Caflisch
    Email Address : atlantis@bioc.unizh.ch
    Institution   : Biochemical Institute of the Universitiy of Zurich
    Date          : July 28, 2001

    The SASA model is a fast implicit solvation model suited for
peptides. It is based on the solvent accessible surface area, using only
two solvation parameters. It approximates this area with a simple
analytical function that is easily derivable. It accounts for the
screening effect of solute charges by using a distant dependent
dielectric function and by neutralizing the charged groups of polar
amino acids in exactly the same way as it is done in the EEF1 model.
The SASA model has been successfully applied to peptides, removing the
major artefacts of in vacuo simulations and even reproducing reversible
folding. Latest benchmarks indicate that a simulation with SASA is only
about 50% slower than an in vacuo simulation.

Files Added:
source/fcm/sasa.fcm
      /misc/sasene.src, sasini.src

Files Modified:
source/charmm/charmm_main.src, iniall.src
      /energy/energy.src, eutil.src, printe.src
      /fcm/energy.fcm

New Documentation:
doc/sasa.doc

New Testcase:
test/c29test/sasa.inp

  2. c29a2 Change Log Previous Next Top

    The following are introduced and modified during the development period
of August 15, 2001 through February 15, 2002.

2.1 TSRI Developments
    Name          : Mike Crowley, Charles Brooks
    Email Address : crowley@scripps.edu, brooks@scripps.edu
    Institution   : TSRI
    Date          : December 27, 2001

(1) Added OPLS-AA protein force field to CHARMM.
Pref keyword: OPLS
Files modified:
install.com
source/charmm/iniall.src
      /energy/enefscal.src
      /util/prpref.src

(2) Cray Binary File Read
     Modified traj read to permit reading Cray binary files on other machines.
Pref keyword: T3ETRAJ
Files modified:
source/dynamc/dynio.src

(3) Zero Length Message
     Changes to eliminate communication of zero length messages.
Files modified:
source/machdep/paral3.src, parset.src

(4) Fixed IO control in solana
Files modified:
correl/solana.src

(5) Code modifications for serial compile with RCFFT
Pref keyword: RCFFT
Files modified:
install.com
source/nbonds/pmeutil.src

(6) Fixed ##If def logic.
Files modified:
source/mc/mcener.src, mvrtrn.src 

(7) Fixed/added selection logic to quick command.
Pref keyword: QUICK_SELECT
Files modified:
install.com
source/charmm/iniall.src
      /misc/quicka.src
      /util/prpref.src
test/c29test/quicktst.inp [New]
doc/miscom.doc

(8) Changed pref keyword to unique identifier.
Pref keyword: SASAE
Files modified:
install.com
source/charmm/charmm_main.src, iniall.src
      /energy/energy.src, eutil.src, printe.src
      /fcm/energy.fcm
      /misc/sasene.src, sasini.src
      /util/prpref.src

(9) Fixes for using slow and fast shake in same run, especially parallel.
Switched indices on shake pair list from shkapr(n,2) to shkapr(2,n).
Files modified:
source/dynamc/dcntrl.src, mts.src
      /fcm/shake.fcm
      /gener/update.src
      /manip/corman.src, cshake.src, fsshake.src, shake.src

(10) Added compile-code control to test cases.
Files modified:
test/c20test/hbuildst2.inp, imst2test.inp, rigidst2.inp, sbdtest2.inp
            /st2test.inp
    /c26test/block4.inp
    /c29test/sasa.inp

------------------------------------------------------------------------------

2.2 Cornell University Developments
    Name          : Nilesh Banavali, Wonpil Im, Benoit Roux
    Email Address : nilesh.banavali@cornell.edu, wonpil.im@cornell.edu,
                    benoit.roux@med.cornell.edu
    Institution   : Weill Medical College of Cornell University
    Date          : December 27, 2001

(1) Enhancements of Extended Electrostatics, PERT, PBEQ and GSBP
     PBEQ and GSBP are updated and GSBP is implemented within PERT.
Extended electrostatics has been made compatible with PERT.  Substantial
changes were needed in nbondg.src file to make EXTENDED ELEC compatible
with PERT (storage of constant fields, etc...).

Files Modified:
source/fcm/pert.fcm
      /pert/pert.src, epert.src
      /nbonds/nbutil.src, nbondg.src, nbonds.src
      /misc/pbeq.src, gsbp.src, gsbp2.src

(2) Branched IF Command
     The IF command is enhanced to handle IF-THEN-ELSE-ENDIF constructs.

File Modified:
source/charmm/miscom.src

------------------------------------------------------------------------------

2.3 Updates of the SASA Implicit Solvation Model
    Name          : Urs Haberthuer, Joannis Apostolakis and Philippe Ferrara
    Email Address : atlantis@bioc.unizh.ch
    Institution   : Biochemical Institute of the Universitiy of Zurich
    Date          : 26.12.2001

     SASTRM has been moved from the CEPROP array to the CETERM array.
Furthermore, the name to access to SASA mean solvation energy within the
CHARMM input file has been changed from SASA to SASL in order to avoid
confusion with the name used in the getpref() subroutine. The testcase
now works fine.
     Flexibility for the user has been improved.  The assignement of the
SASA parameters is now based on the CHARMM atom types.

Files Modified:
source/energy/energy.src, eutil.src, printe.src
      /fcm/energy.fcm, sasa.fcm
      /misc/sasene.src, sasini.src
doc/sasa.doc
test/c29test/sasa.inp

------------------------------------------------------------------------------

2.4 GAMESS-UK Interface Updated
    Name          : Paul Sherwood
    Email Address : p.sherwood@daresbury.ac.uk
    Institution   : CLRC Daresbury Lab.
    Date          : January 16, 2002

     Makefiles are updated for proper compilation with CHARMM.  Documentation
and testcase are also updated.

Files modified:
install.com
build/UNX/Makefile_alpha, Makefile_alphamp, Makefile_gnu, Makefile_hpux,
         /Makefile_ibmrs, Makefile_ibmsp, Makefile_ibmsp3, Makefile_sgi,
         /Makefile_sun
doc/flucq.doc, gamess-uk.doc
source/flucq/fluqqmmm.src
      /fcm/gamess.fcm
      /gukint/gukini.src
test/cquantumtest/alanine_guk.inp
    /data/alanine_guk.in

------------------------------------------------------------------------------

2.5 Karolinska Institute Fixes and Enhancements
    Name          : Lennart Nilsson
    Email Address : Lennart.Nilsson@biosci.ki.se
    Institution   : Karolinska Institute
    Date          : January 16, 2002

(1) A module for analysis of NMR dispersion relaxation to solvent is being
added.
File modified:
source/correl/solana.src

(2) Allow for, and test, the use of heap expansion on I8-pointer machines
(alpha).  The heap is still addressed using I4 integers.
File modified:
source/machdep/space.src

(3) Check if images are present when PERT is invoked and warn user that the
crystal/image facility has to be invoked AFTER pert.
File modified:
source/pert/pert.src

(4) Document how to run parallel jobs under MPI-implementations not allowing
rewind on stdin.
File modified:
doc/parallel.doc

(5) In many testcases the test for parallel vs non-parallel execution is
changed from

     if ?parallel .eq. 1 then stop
to
     if ?NUMNODE  .gt. 1 then stop

so that testcases are run if an executable with parallel enabled is used
on a single CPU.
Files modified:
test/c20test/constest.inp, enbtest.inp, hbuildst2.inp, ictest.inp
            /imst2test.inp, langtest1.inp, langtest2.inp, powelltes.inp
            /quasi.inp, rigidst2.inp, sbdtest2.inp, st2test.inp
            /vibpafl.inp, vibran.inp, vibrtst.inp, vibwat.inp
    /c22test/covaritst.inp, icfix.inp, icpert.inp, mewtest.inp
            /slowgr.inp, stdtest.inp, travel.inp, window.inp
    /c23test/nmrtest1.inp, nose1.inp, nose2.inp
    /c24test/block3.inp, dimb1.inp, dimb2.inp, pert2.inp, st1.inp
    /c25test/cwat.inp, ewald_grp.inp, ewald_pert.inp, fastest.inp
            /helix.inp, hrbestfit.inp, quiet.inp, rpath1.inp, vibwat25.inp
    /c26test/lambda_dynam.inp, luptst.inp, pathint.inp, tmc.inp, whamtest.inp
    /c27test/ace1.inp, adumb-enum.inp, adumb-phichi.inp, cubesgb.inp
            /gb_cff95_prot-test.inp, lnx_img.inp, testbycc1.inp
            /testbycc2.inp, testeef1.inp, testeef1imag.inp
    /c28test/gb_fep.inp, gb_ldm.inp, gb_mcmd.inp, gbmf19.inp, ldm_lang.inp
            /ldm_lang_slow.inp, ldm_rst.inp, ldm_rst2.inp, ldm_rst2_slow.inp
            /ldm_rst_slow.inp, mcmd_rst.inp, mcmd_rst2.inp, mcmd_rst2_slow.inp
            /mcmd_rst_slow.inp, pssp3.inp, replica_genborn1.inp
            /replica_genborn2.inp, replica_genborn3.inp
    /c29test/sasa.inp
    /ccfftest/cffdyn1.inp

------------------------------------------------------------------------------

2.6 The SPASIBA Force Field
    Name          : P. Lagant, G. Vergoten, R. Stote
    Email Address : rstote@chimie.u-strasbg.fr
    Institution   : Universite Louis Pasteur, CNRS - ISIS 
    Date          : January 28, 2002

The SPASIBA force field is derived from spectroscopic studies.  In the
current implementation, the van der Waals and electrostatic interactions 
of the CHARMM force field are combined with the Urey-Bradley Shimanouchi terms 
for bond stretching, valence angle bending, torsional and improper torsional
internal terms derived from spectroscopic data such as vibrational wavenumbers,
dipole moments, rotation barriers for the SPASIBA force field.

Files Added:
source/gener/setup_spa.src
      /energy/ebaspas.src
      /fcm/spa_code.fcm, spacod.fcm, spapar.fcm

Files Modified:
source/charmm/iniall.src
      /io/coorio.src, parmio.src, psfres.src
      /gener/update.src
      /energy/ecnstr.src, ediff.src, eintern.src,  energy.src, intere.src
      /manip/intcor2.src
      /image/eimage.src
      /fcm/dimens.fcm, psf.fcm, param.fcm, exfunc.fcm
doc/energy.doc

New Testcase:
test/c29test/spasibatest.inp

------------------------------------------------------------------------------

2.7 ACE2 - ACE Enhancements
    Name          : Michael Schaefer
    Email Address : schaefer@piaf.u-strasbg.fr
    Institution   : Universite Louis Pasteur, Strasbourg, France
    Date          : Jan 30, 2002

ACE2 is an improved version of the ACE implicit solvent potential.
It is designed to work for proteins of arbitrary size, as opposed
to the original version of ACE which was limited to polypeptides
of up to 15 residues. This is achieved by an improved treatment
of the Born solvation radii.

Currently, ACE2 appears to work quite reliably in conjunction
with param19, but less so for all-atom parameters.

Atom volumes are now read from file/stdin, and no longer from a volume
table that is appended at the parameter file. Parameter files with
new volumes (Schaefer et al., J. Comp. Chem. 22 (2001), 1857-1879)
compatible with param19, par_all22_prot, par_all27_na are provided.

The new ACE and ACE2 energy routine treats distances between atoms
in the nonbonded exclusion list as invariant.

The ACE energy of the new implementation does NOT reproduce the
energy as calculated by the old (ACE1) implementation!

Files Added:
source/io/aceio.src
toppar/acepar19.inp, acepar22_prot.inp, acepar27_na.inp

Files Removed:
source/nbonds/acean.src
test/c27test/ace2.inp
    /data/param19-1.2.inp

Files Modified:
source/charmm/charmm_main.src, iniall.src
      /dynamc/dcntrl.src, dynlng.src, dynlngv.src
      /fcm/code.fcm, contrl.fcm, inbnd.fcm, param.fcm
      /gener/psfsum.src
      /io/mainio.src, parmio.src
      /manip/scalar.src
      /mc/mc.src, mvhmc.src
      /nbonds/ace.src, enbond.src, nbonds.src, nbutil.src
      /pert/block.src
      /util/selcta.src

Modified Documentation:
doc/ace.doc

New/Modified Testcase:
test/c27test/ace1.inp
    /c29test/ace_v2.inp, ace-adumb.inp
    /data/acepar19.inp

Bugfixes:
(1) ACE and ACE2 work properly when using BLOCK (including
    use of BLOCK while fixing atoms).
(2) changed printout of SIGMA in nbutil, now given in cal/mol/A^2
    i.e., a factor of 1000. bigger.
(3) added IF statement in IESFIX to calculate fixed atom pair
    interaction only once for all nodes (parallel code).

Known bugs:
ACE2 distorts Trp sidechains ("butterfly") when minimizing with param19.
ACE2 may lead to partial unfolding when performing a Langevin dynamics
simulation of a protein with all-hydrogen parameters.
The MC module is not expected to make use of the ACE2 enhancements.

------------------------------------------------------------------------------

2.8 Bugfixes

(1) MC with HMC moves and Tsallis weighting (IACCept 2)
    Name          : Aaron R. Dinner
    Email Address : dinner@gold.cchem.berkeley.edu
    Institution   : University of California, Berkeley
    Date          : 17 January 2002

     The Tsallis acceptance criterion applied to Hybrid Monte Carlo
moves transforms the kinetic energy incorrectly.  c29test/mctest05.inp
checks this problem.  DEKIN and DEPOT are passed separately to TSALLF
via MCACPT (changing the calling sequences to both those routines).
A flag (QHMC) is employed to force the correct operations in TSALLF.

File fixed:
source/mc/mc.src

..............................................................................

(2) Velocity Verlet Restart
    Name          : Arjan van der Vaart
    Email Address : arjan@schubert.u-strasbg.fr
    Institution   : ISIS
    Date          : November 8, 2001

     Restart runs using the velocity Verlet integrator always give SHAKE
errors, when heavy atoms bound to the SHAKE-able hydrogens are fixed.

File fixed:
source/dynamc/dynamvv.src

..............................................................................

(3) Parallel EEF1 Fix
    Name          : Arjan van der Vaart
    Email Address : arjan@schubert.u-strasbg.fr
    Institution   : ISIS
    Date          : December 5, 2001

     The parallel implementation of EEF1 gives wrong solvation (and total)
energies if more than one processor is used (forces are fine).
(Note that the "dirty trick" comments have been deleted, since the
"dirty trick" has been replaced by new code)

File fixed:
source/misc/eef1.src
  3. c29b0 Change Log Previous Next Top

    The following are introduced and modified during the development period
of April 15, 2002 through July 22, 2002.

3.1 Harvard Developments
    Name          : Robert Petrella
    Email Address : petrella@tammy.harvard.edu
    Institution   : Harvard University
    Date          : July 02, 2002

(1) Poisson-Boltzman phi averages
     New commands allow calculation of averages of the electrical potential
over the grid with or without an atom selection.

Files Modified:
source/charmm/iniall.src
      /misc/pbeq.src, pbeq2.src
      /fcm/pbeq.fcm
doc/pbeq.doc

(2) BYCC Non-bonded update
     Exclusion table is simplified.

Files Modified:
source/nbonds/nbndcc.src, nbonds.src, nbutil.src, nbexcl.src
      /fcm/exclar.fcm

(3) Structural correlations in adaptive umbrella sampling 
     New commands in adumb allow calculation of running averages for
various structural variables as a function of the reaction coordinates
of the system.

Files Modified:
source/adumb/cmds.src, umb.src
      /charmm/iniall.src
      /energy/energy.src
      /manip/intcor2.src
      /fcm/umb.fcm, umbcor.fcm 
doc/adumb.doc

------------------------------------------------------------------------------

3.2 Analytical Second Derivatives of the EEF1 Potential
    Name          : Ioan Andricioaei
    Email Address : andricio@tammy.harvard.edu
    Institution   : Harvard University
    Date          : 9 May 2002

    The analytical expression of the second derivative matrix of the EEF1
potential term has been added in.  The main code (i.e., the actual expression
of the analytical second derivatives) was added in subroutine EEF1EN() of
the misc/eef1.src code.  The argument list of the EEF1EN() function has been
modified.  Three extra variables (last three in the argument list) are now
passed: DD1, STACK(IUPT), and QSECD.  In energy/energy.src, Robert 
Petrella's ATOMON, which was being pushed on the stack before, now is 
passed pushed on the heap.

The other code that calls EEF1EN is in the additional three files listed
below, and the EEF1EN() calls of each of them has been changed accordingly,
i.e., by adding the three extra arguments.

Files Modified:
source/misc/eef1.src
      /energy/energy.src, intere.src
      /mc/mcener.src
doc/eef1.doc

New Testcase:
test/c29test/eef1_2ndderiv.inp

------------------------------------------------------------------------------

3.3 Bugfixes for the MC Module
    Name          :  Aaron R. Dinner
    Email Address :  dinner@gold.cchem.berkeley.edu
    Institution   :  University of California, Berkeley
    Date          :  17 June 2002

(1) MC uses its own routines to calculate the ACE energy term, and these
were not updated with the recent change to the ACE energy term.
The fix provides support for both the new ACE1 and ACE2.  The testcase
mctest06.inp was modified accordingly.

(2) The algorithm for picking random displacements within a unit sphere was
changed.  The old algorithm gave a non-uniform distribution in volume (but
did not violate detailed balance, so old simulation results are OK).  This 
modification affects all RTRN type moves.

(3) Some dynamically allocated arrays were not freed properly and now are.
Several testcases were modified to test the memory usage.

(4) Double-comma in one call of LNGFIL is fixed.

Files involved:
source/mc/mc.src             (1, 2, 3 above)
          mcace.src          (1, 2)
          mcener.src         (1)
          movead.src         (3)
          mvhmc.src          (4)
          mvrtrn.src         (3)
test/c29test/mctest03.inp    (2)
             mctest04.inp    (3)
             mctest05.inp    (3)
             mctest06.inp    (1, 3)
             mctest07.inp    (3)

------------------------------------------------------------------------------

3.4 University of Minnesota Developments
    Name          : Mireia Garcia-Viloca and Jiali Gao
    Email Address : garcia@chem.umn.edu
    Institution   : U. of Minnesota
    Date          : July 8, 2002

3.4.1 QM/MM(MOPAC) Enhancements

(1) To allow qm/mm group-group interactions between mm-group and qm-molecule.

     In QM/MM calculations, charges are fully delocalized in the QM region,
making it impossible to separate the QM region into groups.  However, when a
QM molecule is large, greater than a typical cutoff distance, it causes
problems in nonbonded list generation and thus it is desirable to be able to
separate the QM molecule into groups purely for the purpose of QM-MM
nonbonded list.
     The optional keyword [GROUp] is appended to the QUANtum command to allow
this operation.  Note that the GROUp option is purely used for non-bonded
list generation.  Whenever an MM group interacts with any QM group, the MM
group is included in the electronic structure calculation for the entire QM
molecule, i.e., MM-group and QM-molecule interaction.
     Note that the default implementation of QM/MM nonbonded list allows for
construction of the Fock matrix involving QM group - MM group interactions.
This is avoided by specifying the new [GROUp] option.

(2) Reorganization and implementation of the GHO parameters for
    AM1 and PM3 models.

     The GHO parameters for AM1 carbon are moved to qmdata file, and
parameters for the PM3 GHO carbon have been developed and implemented.

(3) Updating AM1 parameters for phosphorus, and PM3 parameters for carbon. 

     MNDO parameters were used in the AM1 model for phosphorus in CHARMM 
previously.  It is corrected in the distribution.  Some GAUSSIAN parameters 
for carbon atom in the PM3 parameter set were missing in the initial
distribution (c28b1) and they are corrected now.

(4) Incorporation of a simple valence bond (SVB) term as a function of 
    a raction coordinate in semiempirical QM/MM calculations.

     An empirical potential is added to the semiempirical  QM/MM potential.
The purpose of including this empirical potential is to correct energetic 
results from low level QM models in order to have good agreement with
experimental or high-level ab initio results on activation energy and
energy of reaction.  The basis of this approach is to assume that 
the low level QM model can give a reasonable description of the shape 
and landscape of the potential energy surface for the problem that is 
being investigated.  The form of the empirical potential function is a 
two state valence bond model and the new keyword in QUANtum [LEPS] is [SVB].

(5) Mulliken population analysis and averaging during a QM/MM MD simulation. 
     A new optional keyword [CHDYn] is included in the QUANtum command to
allow the calculation of average Mulliken population charges along a MD
(QM/MM) trajectory. 

(6) Increase in dimension for EIGS and IFACT in subroutine MULIK2 
     (120 changed to MAXORB) to avoid array overflow. 

(7) Modifications in QM/MM decomposition analysis. 
     A warning message is given when GHO atoms are used (not implemented).
The velocity verlet integrator is now allowed for QM/MM energy decomposition.

Files Modified:
source/fcm/leps.fcm, quantm.fcm
      /quantum/qmdata.src, qmene.src, qmleps.src, qmset.src
      /dynamc/dynamc.src, dynamvv.src
doc/qmmm.doc                                  

..............................................................................

3.4.2 Potential of Mean Force Calculations and CHARMMRATE

(1) To recenter a periodic system to the geometric center of
    the N first residues of the system (generally the protein-substrate
    complex) at each nonbonded list evaluation. 

     Often it is desirable to keep the "active site" of a protein modeled by
periodic boundary conditions near the center of the central unit cell.  In
combination with other facilities, one can avoid the image generation for
periodic boundary systems.  The keyword [CENT NCREs int] has been added as
an optional command under DYNAmics operation.

(2) To rearrange common blocks for biasing potentials in the rxncom.fcm 
    and to change the format for umbrella sampling calculations.

     The array definition and location have been moved from the source file 
to a common block file.

(3) Incorporation of a projection operator of the reaction coordinate in 
    instantaneous normal mode analysis using the CHARMMRATE module. 

     To remove the degree of freedom corresponding to the reaction coordinate
from instantaneous normal mode analysis. The new keyword is [PMFZ] under the 
POLYrate command.

(4) A one line modification of the install.com file for compilation of
    the crate (for charmmrate module) facility with CHARMM-POLYRATE interface.

Files Modified:
source/fcm/charmmrate.fcm, contrl.fcm, rxncom.fcm
      /prate/charmmrate.src
      /dynamc/dcntrl.src
      /nbonds/heurist.src
      /rxncor/rxndef.src, rxnene.src
build/UNX/prate.mk
doc/dynamc.doc, charmmrate.doc

------------------------------------------------------------------------------

3.5 TSRI Developments
    Name          : Scott Brozell, Mike Crowley and Charles L. Brooks, III
    Email Address : crowley@scripps.edu, brooks@scripps.edu
    Institution   : TSRI
    Date          : july 10 2002


(1) Faster image updates

Files Modified:
source/image/upimag.src

(2) Speed enhancing tuning in REWALD aminly for sgi
     pref keyword: FASTEW

Files Modified:
source/charmm/iniall.src
      /nbonds/enbond.src, ewald.src
      /util/prpref.src

(3) Faster communications for fft transpose (parallel)
     pref keyword: MPIFFT

Files Modified:
source/nbonds/pmeutil.src

(4) Long Range vdW
     Continuum model with homogenenous density of atom types
is used to compute the vdW correction to the energy for
interactions beyond the cutoff. THe virial is also corrected.
     pref keyword: LRVDW

Files Modified:
install.com
source/charmm/iniall.src
      /energy/energy.src, printe.src
      /fcm/deflts.fcm, energy.fcm, fast.fcm, inbnd.fcm
      /io/parmio.src
      /nbonds/nbexcl.src, nbutil.src
      /util/prpref.src
doc/nbonds.doc, developer.doc

New Testcase:
test/c29test/lrvdw.inp

(5) NOSKULL option added
     NOSKULL in pref.dat eliminates skull
     pref keyword: NOSKULL

Files Modified:
source/machdep/machutil.src


(6) Bug fixes

Files Modified:
source/machdep/cstuff.src
      /manip/shake.src
test/c29test/mctest03.inp, mctest04.inp

------------------------------------------------------------------------------

3.6 Karolinska Institute Fixes and Enhancements
    Name          : Lennart Nilsson
    Email Address : Lennart.Nilsson@biosci.ki.se
    Institution   : Karolinska Institute
    Date          : June 28, 2002

(1) The ECHO command
     A simplified command for text&data printing has been added. Similar to
"write title" but simpler syntax. 

     [ECHU integer | ECHU]
     ECHO ......

ECHO simply prints remainder of commandline, with standard substitutions,
to OUTU or some other unit as specified by the ECHUnit command (if no unit
number is given, printing will be to OUTU)

Files Modified:
source/charmm/iniall.src, miscom.src
      /fcm/stream.fcm
doc/miscom.doc

(2) New COOR ANAL options
2     New analysis options have been added to COOR ANAL: Analysis of magnetic
relaxation dispersion from solute sites to water protons and Distance
dependent Kirkwood factor (Hochtl et al JCP 109, 4927).

Files Modified:
source/correl/solana.src
doc/corman.doc

(3) Message Clarification
     The restart file written after an "ENERGY CHANGE TOLERANCE EXCEEDED"
error is NOT a regular restart file.  A message about this is now written
to the output file.

Files Modified:
source/dynamc/dynamc.src

(4) RMSD Matrix
     2D-projection of the rmsd matrix computed in RMSDyn has been added
(M.Levitt JMB 168, 621 (1983)).  Output of the RMSD-matrix in matrix format
has been added.  Output from all the rmsd calculations is suppressed unless
PRNLev is increased
(This command should ideally access the trajectories through READCV;
this however requires modifications to allow READCV to handle multiple
trajetories, which I have not finished yet.)

Files Modified:
source/manip/rmsdyn.src
      /util/imsl.src (error output from zxmin has been suppressed)
doc/dynamc.doc, subst.doc
test/c29test/rmsdyn.inp  (New)

(5) COOR DMAT Clarification
     COOR DMAT output in either the old "full" format or the matrix-elements
only has been cleanly separated.  An option to write out a matrix with a "1"
as the element when a contact is present, and "0" otherwise, has been added.
This is useful when the matrix is to be used as a projection matrix later on.

Files Modified 
source/manip/corman2.src
doc/corman.doc

(6) PERT Variables
     Three new CHARMM variables with results from PERT-runs have been added

Files Modified
source/pert/pert.src
doc/subst.doc

(7) I/O Control
     I/O statements in MMFP have been protected for parallel runs
(IF(PRNLEV.GE.2))

Files Modified
source/misc/mmfp.src

(8) Mass Unit Correction
     Units for the "mass" of the extended system thermal piston have been
corrected in pressure.doc

Files Modified
doc/pressure.doc

------------------------------------------------------------------------------

3.7 NIH Developments
    Names         : Milan Hodoscek, Lee Woodcock, Paul Sherwood, Bernie Brooks
    Email Address : milan@cmm.ki.si
    Institution   : National Institutes of Health
    Date          : July 9, 2002

3.7.1  QM/MM extension of REPLICA/PATH and similar methods

    We extended the functinality of the REPLICA/PATH like methods to enable
running them with the QM/MM potential. All the functionality of the RPATH
command is implemented. The method uses parallel/parallel setup.

Files Modified:
source/cadint/cadini.src
       /dynamc/dynamcv.src
       /fcm/gamess.fcm, parallel.fcm
       /flucq/flucq.src, fluqqmmm.src
       /gamini/ddi.src
       /gener/update.src
       /gukint/gukini.src
       /machdep/paral1.src, paral2.src, paral3.src, parset.src
       /nbonds/nbonda.src, nbonds.src
       /quantum/qmset.src
doc/replica.doc

..............................................................................

3.7.2 Merging of GAMESS, GAMESS-UK, and DIESEL interfaces into a single file

    The interfaces to the programs GAMESS, GAMESS-UK, and DIESEL were merged.

Files Modified:
source/gukint/gukini.src
      /charmm/charmm_main.src
      /energy/energy.src
      /fcm/gamess.fcm
doc/gamess.doc
build/UNX/gmscomp

New files:
doc/diesel.doc

Files Deleted:
source/gamint/gamini.src

------------------------------------------------------------------------------

3.8 GAMESS-UK Interface Updated
    Name          : Paul Sherwood
    Email Address : p.sherwood@daresbury.ac.uk
    Institution   : CLRC Daresbury Lab.
    Date          : July 12, 2002

3.8.1 INTEGER8 pref.dat keyword

     This provides a generic way to switch to 64 bit integers, it is a
generalisation of the code provided for GNUALPHA by Ben Webb with a few
additional changes that were found to be necessary.
     It has been tested together with the sgi64 and alpha targets
it should also work with gnu/cfort (the original GNUALPHA platform)

Files Modified:
install.com
source/machdep/paral1.src, paral2.src, paral3.src, startup.src, cstuff.src
build/UNX/Makefile_alpha, Makefile_sgi64

..............................................................................

3.8.2 Other GAMESS-UK Interfacing changes

     Added tests on GAMESSUK pref.dat keyword to remove BLAS code when
linked to GAMESS-UK (will be provided either by GAMESS-UK itself or by
libraries it will use).

File Modified:
source/util/array.src
      /gukint/gukini.src

------------------------------------------------------------------------------

3.9 Bugfixes and Enhancements

(1) National Institutes of Chemistry (Slovenia) Bugfixes
    Name          : Milan Hodoscek
    Email Address : milan@cmm.ki.si
    Institution   : National Institute of Chemistry, Ljubljana, Slovenia
    Date          : July 9, 2002

(1.1) Compilation on HP, and others

    Missing stream.fcm for prnlev variable (HP didn't compile) in
efscalar_cff.fcm, missing comma in format statement.
    
Files involved:
source/cff/efscalar_cff.src
      /pert/block.src

..............................................................................

(1.2) Write the trajectory file in the ASCII double precision formart

    This enhancement is needed for easier analysis of integration algorithms
and others.

Files modified:
source/charmm/iniall.src
      /dynamc/dcntrl.src, dynio.src
      /fcm/reawri.fcm
      /minmiz/minmiz.src

..............................................................................


(1.3) MD-GRAPE-II modifications

    MD-GRAPE-II support was extended for ewald method.

Files involved:
source/nbonds/grape.src
      /nbonds/nbutil.src
      /dynamc/dynamc.src

..............................................................................

(1.4) Support for general exponent in the HQBM method

    The HQBM method was enhanced to support any exponent.

Files involved:
source/energy/hqbm.src
      /fcm/hqbm.fcm
doc/hqbm.doc

..............................................................................

(1.5) LARGE XXLARGE dimensions generalized

    The dimens.fcm was fixed to support LARGE and XXLARGE

Files involved:
source/fcm/dimens.fcm

..............................................................................

(1.6) PBEQ fix: allows Epsilon other than one with the forces

    Allows Epsilon other than one with the forces

Files involved:
source/misc/pbeq2.src

..............................................................................

(1.7) Subroutine name conflict with GAMESS

    Subroutines DIVIDE and FACT have the same names as in the GAMESS program

Files involved:
source/nbonds/fma.src
      /util/calc.src

..............................................................................

(2) Installation of a SUN MPI version
    Name          : Roland H. Stote
    Email Address : rstote@chimie.u-strasbg.fr
    Institution   : Universite Louis Pasteur, CNRS - ISIS 
    Date          : June 18, 2002

     A SUN MPI port is added.  install.com is modified to set up a separate
subdirectory for a SUN MPI version of CHARMM.  The options used in
Makefile_summpi are valid as well for SUN monoprocessor.

Files involved:
install.com
build/UNX/Makefile_sunmpi (New)

..............................................................................

(3) TSM bugfix
    Name          : Robert Petrella
    Email Address : petrella@tammy.harvard.edu
    Institution   : Harvard University
    Date          : July 02, 2002

     Format I is fixed to I4.

File Fixed:
source/pert/tsms.src

..............................................................................

(4) Array Declared for the Grid Code
    Name          :  Aaron R. Dinner
    Email Address :  dinner@gold.cchem.berkeley.edu
    Institution   :  University of California, Berkeley
    Date          :  17 June 2002

     GridRadii is a local array but declared adjustable size in GridEnergy.
The bug prevents compilation on at least the Portland Group compilers.
The GridRadii array was not used by GridEnergy and was deleted.

File Fixed:
source/misc/grid.src 

..............................................................................

(5) SHAKE Compile
    Name          : Lennart Nilsson
    Email Address : Lennart.Nilsson@biosci.ki.se
    Institution   : Karolinska Institute
    Date          : June 28, 2002

     nconst and nconst_tot were defined twice in one routine.
Subroutine fill_fstprs seems to require MPI communications: added
protection by ##IF MPI

File Fixed:
source/manip/shake.src


..............................................................................

(6) NULL String
    Name          : Lennart Nilsson
    Email Address : Lennart.Nilsson@biosci.ki.se
    Institution   : Karolinska Institute
    Date          : June 28, 2002

     Replaced no-good null string ('') in subroutine calls.  This affects
the testcase output of c29test/sasa.inp.2

File Fixed:
source/misc/sasini.src

..............................................................................

(7) PBEQ Probe Size
    Name          : Benoit Roux
    Email Address : Benoit.Roux@med.cornell.edu
    Institution   : Weill Medical College of Cornell University
    Date          : July 4th 2002

     If the user is requesting a re-entrant calculation of the solvent-
accessible surface area of the protein, then the water probe should be
1.4 angtroms.  Otherwise the result is not meaningful.
Therefore, we modified the setup for the default values so that a probe size
of zero is used, unless a re-entrant calculation is requested.
More information about this is also printed out to warn the user.

File Fixed:
source/misc/pbeq.src

  4. c29b1 Change Log Previous Next Top

    Some features are still under development and are determined not to be
included in the forthcoming beta releases of CHARMM 29 at the CHARMM meeting
(Harvard University, July 12-13, 2002).  The c29b1 release of August 15, 2002
is prepared with the following functions removed.

4.1 The SPASIBA Force Field Removed

Files Removed:
source/gener/setup_spa.src
      /energy/ebaspas.src
      /fcm/spa_code.fcm, spacod.fcm, spapar.fcm
test/c29test/spasibatest.inp

Files Modified:
source/charmm/iniall.src
      /io/coorio.src, parmio.src, psfres.src
      /gener/update.src
      /energy/ecnstr.src, ediff.src, eintern.src,  energy.src, intere.src
      /manip/intcor2.src
      /image/eimage.src
      /fcm/dimens.fcm, psf.fcm, param.fcm, exfunc.fcm
doc/energy.doc

------------------------------------------------------------------------------

4.2 GBMF, GBMV, and CMAP codes

     Following the request of the original developer of TSRI, the GBMF code
is removed and is replaced by the GBMV code.  The TSRI check-in was integrated
into c29b0 as a whole and GBMF, GBMV and CMAP codes are removed for c29b1.
GBMV and CMAP will be further developed in c30.

Files removed:
source/energy/gbmv.src
      /fcm/gbmv.fcm
test/c28test/gbmf19.inp

Files Modified (GBMF/GBMV):
install.com
source/charmm/charmm_main.src, iniall.src
      /energy/energy.src, genborn.src, intere.src
      /fcm/genborn.fcm
      /io/psfres.src
      /util/prpref.src
doc/developer.doc

Files Modified (CMAP):
install.com
source/charmm/charmm_main.src, iniall.src
      /correl/correl.src
      /dynamc/dcntrl.src, dynamln.src, dynio.src, mts.src, trnphi.src
      /energy/ecntrl.src, ediff.src, eintern.src, energy.src, eutil.src
             /intere.src, printe.src
      /fcm/block.fcm, code.fcm, dimens.fcm, energy.fcm, exfunc.fcm
          /image.fcm, param.fcm, pert.fcm, psf.fcm, rtf.fcm, tbmtsp.fcm
          /tsmh.fcm
      /flucq/flucq.src
      /gener/genpsf.src, modpsf.src, psfsum.src, replica.src, update.src
      /image/eimage.src, images.src, imagio.src, upimag.src
      /io/mainio.src, psfres.src, rtfio.src
      /manip/cstran.src
      /mbond/mbond.src
      /mc/mc.src, mcener.src
      /minmiz/egrad1.src, nraph.src
      /misc/hbuild.src, testch.src, xray.src
      /pert/block.src, epert.src, icfcnf.src, icpert.src, pert.src
           /tsme.src, tsms.src
      /quantum/qmset.src
      /util/prpref.src, sort.src
      /vibran/vibran.src, vibsub.src

------------------------------------------------------------------------------

4.3 Structural Overlap Code Remvoed

     Structural overlap method by using Gaussian overlap integrals was
integrated into c29b0 with Milan's other check-in and is removed for
c29b1.  The OVERLAP code will be developed in c30.

Files Removed:
source/misc/olap.src
      /energy/eolap.src
      /fcm/olap.src

Files Modified:
source/charmm/charmm_main.src, iniall.src
      /energy/energy.src, eutil.src, printe.src
      /fcm/energy.fcm
      /util/prpref.src

  5. c29b2 Change Log Previous Next Top

    The following are introduced and modified during the development period
of August 15, 2002 through February 15, 2003.

5.1 TSRI Developments
    Name          : Sandeep Patel, Tom Cleveland, Z. Zhou, Mike Crowley,
                    Charles L. Brooks, III
    Email Address : crowley@sripps.edu, brooks@scripps.edu
    Institution   : TSRI
    Date          : January 7, 2003

5.1.1 Makefiles and Code for Recent IBM and ALPHA machines

     The modification involves better Makefiles and code changes for
more recent IBM machines, alpha and multiple MPI libraries (MPICH,
LAM, MPI_LITE, vendor).

     For ALPHA platforms, modern alpha compile flags are incorporated
and alphamp is removed.

     ibmrs, ibmsp, ibmsp3 are replaced with ibmaix and ibmaixmp in 
install.com.  The pref keyword used is IBMAIX.  This should work on
all current ibm machines and allow us to remove the other IBM stuff
from code and from install.com.  For now, the only ibm architectures
that install.com will allow are ibmaix for serial and ibmaixmp for
parallel with vendor mpi.

Files Added:
build/UNX/Makefile_ibmaix, Makefile_ibmaixmp

Files Modified:
install.com
test/test.com
build/UNX/Makefile_alpha
source/fcm/impnon.fcm, machio.src, machutil.src, space.src, startup.src

..............................................................................

5.1.2 FASTEW for SGI

     Loop rearrangement and temp array usage for speedup in REWALD().
Only applies to sgi.  FC3 macro is set to -O3 in Makefile_sgi and
Makefile_sgi64.

Files Modified:
install.com
build/UNX/Makefile_sgi, Makefile_sgi64, nbonds.mk
source/fcm/ewald.fcm
      /nbonds/ewald.src

..............................................................................

5.1.3 Testcase Enhancements

     Tests are changed to give consistent message when they are not
performed.  Exit via a stream of notrun.inp, which puts a string that
can be detected with grep.  This method keeps the excess verbiage out
of the test input file and makes the message immune to typos.

File Added:
test/notrun.inp

Files Modified:
test/c20test/constest.inp, enbtest.inp, hbuildst2.inp, imst2test.inp
            /langtest1.inp, langtest2.inp, maatest.inp, powelltes.inp
            /quasi.inp, rigidst2.inp, sbdtest2.inp, st2test.inp
            /vibpafl.inp, vibran.inp, vibrtst.inp, vibwat.inp
test/c22test/block1.inp, block2.inp, covaritst.inp, icfix.inp
            /icpert.inp, mewtest.inp, path.inp, slowgr.inp
            /stdtest.inp, travel.inp, umbrella.inp, window.inp
            /xtlala2.inp
test/c23test/clustst.inp, mtsm1.inp, mtsm2.inp, nmrtest1.inp, nose1.inp
            /nose2.inp, replica.inp, rism.inp
test/c24test/block3.inp, dihtest1.inp, dimb1.inp, dimb2.inp, dyn4dtest.inp
            /pert2.inp, tntest1.inp
test/c25test/cortst25.inp, cwat.inp, helix.inp, mtsm3.inp, pull.inp
            /resdtest.inp, rpath1.inp, vibwat25.inp
test/c26test/block4.inp, cftigas.inp, cftmgas.inp, conmin.inp, lambda_dynam.inp
            /lambda_energy.inp, luptst.inp, mbtest18.inp, mbtest19.inp
            /mbtest24.inp, mbtest25.inp, pathint.inp, pbeqtest1.inp
            /pbeqtest2.inp, pm6test1.inp, tga.inp, tgall.inp, tip4ptest.inp
            /tmc.inp, whamtest.inp
test/c27test/ace1.inp, adumb-enum.inp, adumb-phichi.inp, aspenr.inp
            /gb_cff-data.inp, gb_cff95_na-test.inp, gb_cff95_prot-test.inp
            /gb_mmff94_na-test.inp, gb_mmff94_prot-test.inp, gb_mmff94_test.inp
            /lnx_img.inp, mbtest24.inp, mbtest25.inp, mbtest28.inp
            /mbtest29.inp, mbtest30.inp, mctest02.inp, pbeqtest3.inp
            /testbycc1.inp, testbycc2.inp, testeef1.inp, testeef1imag.inp
test/c28test/gb_block.inp, gb_fep.inp, gb_ldm.inp, gb_mcmd.inp
            /grid_2.inp, gridtest.inp, ldm_lang.inp, ldm_lang_slow.inp
            /ldm_rst.inp, ldm_rst2.inp, ldm_rst2_slow.inp, ldm_rst_slow.inp
            /mcmd_rst.inp, mcmd_rst2.inp, mcmd_rst2_slow.inp, mcmd_rst_slow.inp
            /pssp1.inp, pssp2.inp, pssp3.inp, replica_genborn1.inp
            /replica_genborn2.inp, replica_genborn3.inp
test/c29test/ace-adumb.inp, ace_v2.inp, eef1_2ndderiv.inp, mctest03.inp
            /mctest04.inp, mctest05.inp, mctest06.inp, mctest07.inp
            /sasa.inp
test/ccfftest/cffdyn1.inp
test/clongtest/gsbptest1.inp, gsbptest2.inp, mctest01.inp, pbeqtest5.inp
              /pbeqtest6.inp, pbeqtest7.inp
test/cquantumtest/alanine_guk.inp, cr01.inp, cr02.inp, fqam1.inp
                 /fqgamess.inp, fqguk.inp, gmstst.inp, pm3.inp
                 /qmfep.inp, quantum1.inp, testleps.inp

------------------------------------------------------------------------------

5.2 ELSEIF Fix
    Name          : Benoit Roux
    Email Address : benoit.roux@med.cornell.edu 
    Institution   : Weill Medical College of Cornell University
    Date          : December 31, 2002

     There was a little screwup in the nested if-else-endif structure.
The nested if-else-endif structure was messing up the original if
statement of miscom.  The fix implements complete backward compatibility.

File involved:
source/charmm/miscom.src

Top