--------------------------------------------------------------------------- CHARMM22.0.b Release April 22, 1991 CHARMM22.0.b1 Release September 30, 1991 CHARMM22 Release January 1, 1992 c22g1 Release February 15, 1992 c22g2 Release July 7, 1992 CHARMM23.0 c23a1 Developmental August 15, 1992 c23a2 Developmental October 25, 1992 c23f Developmental March 1, 1993 c23f1 Developmental March 15, 1993 c23f2 Developmental August 15, 1993 c23f3 Release February 1, 1994 c23f4 Release August 15, 1994 c23f5 Release March 15, 1995 CHARMM 24.0 c24a1 Developmental February 15, 1994 c24x1 Evaluation February 15, 1994 c24a2 Developmental August 15, 1994 c24a3 Developmental March 15, 1995 c24b1 Release August 15, 1995 c24b2 Release February 15, 1996 c24g1 Release August 15, 1996 c24g2 Release February 15, 1997 CHARMM 25.0 c25a0 Developmental August 15, 1995 c25a1 Developmental February 15, 1996 c25a2 Developmental August 15, 1996 c25a3 Developmental February 15, 1997 c25b1 Release August 15, 1997 c25b2 Release February 15, 1998 CHARMM 26.0 c26a1 Developmental August 15, 1997 c26a2 Developmental February 15, 1998 c26b1 Release September 30, 1998 c26b2 Release February 15, 1999 CHARMM 27.0 c27a1 Developmental September 30, 1998 c27a2 Developmental February 15, 1999 c27b1 Release August 15, 1999 c27b2 Release February 15, 2000 c27b3 Release August 15, 2000 c27b4 Release February 15, 2001 CHARMM 28.0 c28a1 Developmental August 15, 1999 c28a2 Developmental February 15, 2000 c28a3 Developmental August 15, 2000 c28a4 Developmental February 15, 2001 c28b1 Release August 15, 2001 c28b2 Release February 15, 2002 CHARMM 29.0 c29a1 Developmental August 15, 2001 c29a2 Developmental April 15, 2002 c29b1 Release August 15, 2002 c29b2 Release February 15, 2003 CHARMM 30.0 c30a1 Developmental August 15, 2002 c30a2 Developmental February 15, 2003 c30a2x Evaluation February 15, 2003 c30b1 Release August 15, 2003 c30b2 Release February 15, 2004 CHARMM 31.0 c31a1 Developmental August 15, 2003 c31a2 Developmental February 15, 2004 c31b1 Release August 15, 2004 c31b2 Release February 15, 2005 --------------------------------------------------------------------------- [0] About CHARMM31 Development At the July 2003 CHARMM meeting (The Scripps Research Instutute, San Diego, California, July 11-13, 2003), we decided to release CHARMM c30b1 with incorporating enhancements and fixes to the developmental version c30a2. The modified source code and features are described in ChangeLog.c30, [3] c30b1 Change Log. As an on-going project, CHARMM development will be carried out with CHARMM version 31 series. The c31a1 is developed from the expanded version c30a2x. The expansion detail is reported below in the c30a2x Change Log section. The first distribution version is c30b1 and the first developmental version is c31a1 in the c31 development lines. The following describes new features and enhancements incorporated in the c31 development period. =========================================================================== [1] c30a2x Change Log At the July 2002 CHARMM meeting, we planned to prepare a c30-expansion and to have it reviewed by Charles L. Brooks, III, Bernard L. Brooks and Carol B. Post. c30a2x is completed as the expansion of c30a2. 1.1 Expansion of ID names and Allowed Number of Atoms Name : Youngdo Won and Bernard R. Brooks Email Address : won@ihanyang.ac.kr, brb@helix.nih.gov Institution : National Institute of Health Date : February 12, 2003 Objectives ---------- CHARMM is expanded to handle upto ten billion atoms and eight characters for segment ID, residue ID, residue names and atom names. The expansion is implemented in a conservative way that (1) Output and file formats are unchanged when CHARMM can warrant a normal run, i.e. NATOM < 100000, and less than five characters are used for SEGID, RESID, RES and TYPE. (2) The number of atoms (NATOM) and the number of characters used in SEGID, RESID, RES and TYPE are checked when PSF is modified (created) and when an output is made into a file. Only when needed, the expanded formats are used. (3) All previous version data files are accepted. All CHARMM I/O functions are made backward compatible. (4) Expanded format files carry a mark (for example, "EXT") that does not disturb previous version I/O. (5) The MISCellaneous command "IOFOrmat" may enforce either "OLD format" or "EXPanded format". Data File Format ---------------- All data files of the previous versions should be read by the expanded CHARMM. The normal format data files generated by the expanded CHARMM should be read by the previous versions. (1) PSF I/O Expanded format is marked by "PSF EXT" on the header line. Only expanded format files carry "EXT", which signals the expanded format for reading. Expanded format uses I10 atom numbers and eight-character PSF ID names. (2) COORdinate Files Expanded format is marked by "EXT" on the first line after the title lines where the number of atoms is put. "EXT" indicates the expanded coordinate file format. All field widths are doubled: Atom_No Res_No Res_Name Atom_ID X Y Z SEGID RESID W Normal: I5 I5 1X,A4 1X,A4 3F10.5 1X,A4 1X,A4 F10.5 Expanded: I10 I10 2X,A8 2X,A8 3F20.10 2X,A8 2X,A8 F20.10 (3) SEQUence I/O Modified to match the expanded coordinate file format. (4) HBONd, IC, CONStraint, XRAY, IMAGe I/O Atom numbers use the I10 format and PSF ID names use the A format with the (1:idleng) width control. Implementation -------------- Atom numbers are written in I10 and are read in the free format for the expanded format I/O. PSF ID names (SEGID, RESID, RES and TYPE) are written in the "A" format with the width control (1:idleng), where idleng is set to 4 for normal and 8 for expanded I/O. A PSF ID is read with the NEXTA8 function. In psf.fcm CHARACTER*4 SEGID, RESID, RES, TYPE is changed to CHARACTER*8 SEGID, RESID, RES, TYPE Note that CSTACK is also expanded to CHARACTER*8. In stream.fcm, three logical QOLDFMT, QNEWFMT, and QEXTFMT and one integer IDLENG are placed. Defaults are .FALSE. for QOLDFMT and QNEWFMT and QEXTFMT is determined at the time of I/O request. QOLDFMT and QNEWFMT are set by the miscellaneous command IOFOrmat and enforce the normal and the expanded format I/O, respectively. The command syntax is IOFOrmat [EXTEnded ] [NOEXtended] The format (qextfmt) is determined by calling a logical function qxform() (util/string.src), which checks if qextfmt is previously set, if qoldfmt or qnewfmt is set, if NATOM is greater than 100000, and if any PSFID name is in more than five characters. qxform() should be called before writing out a file. I/O of PSFID names are done with the "A" format with (1:idleng). For example, WRITE(OUTU,'(A,A)') 'The Segment ID is ',SEGID(ISEG)(1:idleng) idleng is set to 8 when qextfmt is set (.true.). The default is 4, for the normal I/O. Modified Source: source/adumb/cmds.src, eadumb.src /cadint/cadini.src /cff/enbonda_cff.src /charmm/iniall.src, miscom.src, anacor.src, clustr.src /correl.src, corrio.src, mantim.src, solana.src /emap/emapsubs.src /energy/dmcons.src, ebaspas.src, ediff.src, eintern.src /enst2.src, epull.src, polar.src /fcm/cstack.fcm, exfunc.fcm, heappr.fcm, psf.fcm, stream.fcm /gener/genpsf.src, hbonds.src, mkpres.src, modpsf.src /psfsum.src, replica.src /graphics/drawit.src, grutil.src /gukint/gukini.src /image/cnbndm.src, images.src, imagio.src, nbondm.src, upimag.src /io/coorio.src, mainio.src, psfres.src, univio.src /machdep/quanta.src, space.src /manip/corman2.src, corman3.src, corman.src, cshake.src, cstran.src /fsshake.src, hbanal.src, intcor.src, pucker.src, rgyr.src /rmsdyn.src, scalar.src, shake.src, wrgaus.src /misc/aspener.src, eef1.src, genetic.src, hbuild.src, mmfp.src /nmr.src, noe.src, quicka.src, resdist.src, testch.src, xray.src /mmff/datastruc.src, merckio.src /molvib/molinp.src, molvio.src, molvsb.src /nbonds/cnbnd.src, enbonda.src, ewald.src, nbonda.src, nbondg.src /nbutil.src /pert/epert.src, icfix.src, icpert.src, tsms.src /quantum/addlnat.src, qmset.src /sccdftbint/sccdftbini.src /shapes/mdlio.src /solvation/coorman.src, deriv.src, rismio.src, soluu.src /soluv.src, asolvv.src /util/array.src, chutil.src, selcta.src, string.src, util.src /vibran/quasi.src, vibio.src, vibsub.src Modified Documentation: doc/miscom.doc, io.doc Modified Testcase: test/c24test/pert2.inp ! free format assumes four character names .............................................................................. 1.2 Energy Write Overflow Name : Youngdo Won Email Address : won@ihanyang.ac.kr Institution : Hanyang University Date : February 13, 2003 printe.src is modified to print large energy values in the assigned space. The normal format is F13.5, which is adjusted to F13.4, F13.3, F13.2, F13.1, and E13.5 in order to yield readable numbers. No more "**********" will be seen in the output. File Modified: source/energy/printe.src ============================================================================== [2] c31a1 Change Log The following are introduced and modified during the development period of February 15, 2002 through August 15, 2002. The developments are based on two CHARMM versions: c30a2 and c30a2x. Some enhancements and fixes to c30a2 are also incorporated into c30b1 as described in the c31b1 release note (ChangeLog.c30) and indicated below. 2.1 TSRI Developments Name : Wonpil Im, John Karanicolas, Sandeep Patel, Scott Brozell, Mike Crowley, Charles L. Brooks, III Email Address : wonpil@scripps.edu , crowley@sripps.edu, brooks@scripps.edu Institution : The Scripps Research Institute Date : June 10, 2003 2.1.1. Generalized Born with a simple SWitching (GBSW) The GBSW module provides the (electrostatic + nonpolar) solvation energy and forces. A Generalized Born method is used for the electrostatic part and the solvent-exposed surface ares for the nonpolar part with a phenomenological surface tension coefficient. Based on volume integration schemes used in the GBMV module [M.S. Lee, F.R. Salabury, Jr., and C.L. Brooks III, J. Chem. Phys., 116, 10606 (2002)], we have recast the calculation of the self-electrostatic solvation energy to utilize a simple smoothing function at the dielectric boundary. The GBSW model is formulated in this manner to provide consistency with the Poisson-Boltzmann (PB) theory previously developed to yield numerically-stable electrostatic solvation forces based on finite- difference methods [W. Im, D. Beglov, and B. Roux, Comp. Phys. Comm., 111, 59 (1998)]. However, it is also possible to mimic the PB results with the molecular surface by reparametrizing two adjustable parameters, a_0 to modulate the Coulomb field term and a_1 to include a correction term beyond Coulomb field. The GBSW module takes the influence of biological membranes into account. Consistent with continuum Poisson-Boltzmann (PB) electrostatics, the membrane is approximated as an solvent-inaccessible infinite planar low-dielectric slab. The membrane GB model closely reproduces the PB electrostatic solvation energy profile across the membrane. The GBSW module works with the IMAGE facility. The GBSW calculations are about 4 times slower than the corresponding vacuum calculations. Using the simple smoothing function makes the present GB model roughly 2-3 times faster than the GBMV module. keyword: GBSW Files Added: source/misc/gbsw.src Files Modified: install.com source/charmm/charmm_main.src /energy/energy.src New Documentation: doc/gbsw.doc New Testcase: test/c31test/gbsw.test .............................................................................. 2.1.2 Support for usage of a modified van der Waals term Setting the flag "ETEN" to "ON or OFF" switches the van der Waals to a modified Lennard-Jones function containing an attractive r^-10 term and repulsive r^-12 and r^-6 terms. This was introduced to support simulation of the Go models built by the webserver at mmtsb.scripps.edu/webservices/gomodel.html keyword: GOMODEL Files Modified: source/charmm/charmm_main.src, iniall.src /energy/ediff.src, energy.src /fcm/inbnd.fcm /manip/corman3.src /misc/hbuild.src /nbonds/enbf2.src, enbfast.src, enbond.src, enbonda.src, /enbondg.src, evdwf.src /quantum/qmene.src /util/prpref.src Modified Documentation: doc/energy.doc New Testcases: test/c31test/eten_1.inp, eten_2.inp .............................................................................. 2.1.3 Polarizable model fluctuating charges (c30b1) Code for generalized born model added. Works with proteins. Calling argument list of WRITCV is fixed. Files added: toppar/par_all30_cheq_prot.inp, top_all30_cheq_prot.inp Files Modified: doc/cheq.doc source/cheq/fqener.src /dynamc/dcntrl.src, dynamc4.src, dynamvv.src, nose.src /fcm/cheqdyn.fcm /minmiz/steepd.src /util/prpref.src New Testcases: test/c30test/cq1fsvvacu.inp, cq1fsvwat.inp, cqmeohvacu.inp, cqnmawat.inp /cqtip4pbulk.inp, cq1ab1gbdyn.inp test/data/top_all30_cheq_prot.inp, 1ab1gbtest.crd, 1fsv_min.crd, cubic.xtl .............................................................................. 2.1.4 Build fixes (c30b1) IBM platform makefils are fixed and some compile options are set for platform dependent compilation. Compromise solution to avoid ifc optimization problems for svdcmp.f and intcor2.f. FC1 is used in 1adumb.mk and manip.mk for compilation, respectively, where ifc FC1 is set to -O0. stdlib.h is added in socket.c. IBMAIX key is added in machutil.src. test.com is fixed to run IBM parallel testing. Files modified: build/UNX/Makefile_alpha, Makefile_ibmaix, Makefile_ibmaixmp, /Makefile_ibmsp3 /adumb.mk, manip.mk source/machdep/socket.c, machutil.src test/test.com .............................................................................. 2.1.5 GBMV Fixes (c30b1) New parametrization and bugfixes. Added a guard around a sequence of assignments in Subroutine RunGBMV1. Removed local variable R2 and from Subroutine RunGBMVGrid1. Completely removed local variable RF from Subroutine RunGBMVGrid1. Files modified: source/energy/energy.src, gbmv.src, intere.src /fcm/gbmv.fcm .............................................................................. 2.1.6 Standard coding clean-up (c30b1) Modified SUBROUTINE GAUSSJ to conform to Charmm coding standards (fqener.src). PARALLEL code fixes (epath.src). quicksort routine is moved out of ##IF block for CHEQ (pbeq.src). CHEQ and FASTEW conflict is terminated with an error message (ewald.src). Some print statements removed (tsms.src) and variables are initialized (wham.src). Files modified: source/cheq/fqener.src /energy/epath.src /misc/pbeq.src /nbonds/ewald.src /pert/tsms.src, wham.src .............................................................................. 2.1.7 Testcase fixes (c30b1) fixes for missing parameters (pathint) and to multinode I/O problems. In molvt6, TOLGrad is set to 0.000002 to attain convergence on SGI platforms. Files modified: test/c26test/pathint.inp /c27test/gb_mmff94_na-test.inp, gb_mmff94_prot-test.inp /gb_mmff94_test.inp /c29test/mctest03.inp, mctest04.inp /c30test/cftiangl.inp, cftidihe.inp, cftidist.inp, cftmala10.inp, /cftmtst1.inp, cftmtst2.inp, cftmtst3.inp, cftmtst4.inp /cftmtst5.inp, molvt6.inp ------------------------------------------------------------------------------ 2.2 Cornell University Developments Name : Guillaume Lamoureux and Benoit Roux Email Address : Benoit.Roux@med.cornell.edu Institution : Weill Medical College of Cornell University Date : June 24, 2003 2.2.1 Dipole correlation and polarizability calculations In the CORRel command, the MOVIng keyword can be used to compute the moving average of a time series. In the VIBRAN command, the DIPOle keyword allows the command to compute the molecular polarizability. It prints to contribution of each vibration mode to the polarizability tensor, as well as the total polarizability. It ignores the rotation and translation modes. Files Modified: source/correl/correl.src, anacor.src, mantim.src /vibran/vibio.src Modified Documentation: doc/correl.doc, vibran.doc .............................................................................. 2.2.2 DRUDE command for polarizable force field The DRUDE command generates a polarizable system by modifying the topology and parameters of an existing non-polarizable system. For each selected atom, it creates a "Drude oscillator" by attaching to the atom an additional particle (using a fictitious chemical bond of length zero and of force constant 'KDRUDE = k/2'). Each Drude particle is given a mass and a charge, taken from the mass and the charge of its atom (so that the total mass and charge are conserved for the "atom-Drude" pair). Note: The calling syntax of the CODES subroutine is changed. Files Added: source/gener/drude.src Files Modified: source/fcm/psf.fcm, ssbp.fcm /cff/ewald_cff.src /charmm/charmm_main.src, iniall.src /correl/correl.src /dynamc/trnphi.src /energy/energy.src /gener/genpsf.src, psfsum.src, update.src /manip/shake.src /misc/hbuild.src, ssbp.src /mmff/enbfast_mm.src, enbscalar_mm.src /nbonds/ewald.src, evdwf.src, ewaldf2.src /pert/pert.src, epert.src /quantum/qmset.src New Documentation: doc/drude.doc New Testcases: test/c31test/drude_benzene.inp, swm4.inp .............................................................................. 2.2.3 TPCONTROL and extended dynamics The TPCONTROL command specifies the thermodynamic ensemble to be simulated with "DYNA VV2", using extended dynamics: Nose-Hoover equations for constant volume and constant temperature (the NVT ensemble) and Anderson-Hoover equations for constant pressure and temperature (the NPT ensemble). It allows multiple thermostats. "DYNA VV2" is a velocity-Verlet algorithm created to simulate efficiently the motion of Drude oscillators (created by the DRUDE command), and it understand the special nature of the Drude oscillators. Note: The calling syntaxes of the subroutines WRIDYN and READYN are changed. Files Added: source/dynamc/tpcontrol.src, dynamvv2.src Files Modified: source/fcm/nose.fcm /dynamc/dcntrl.src, dynio.src, dynamc.src /io/coorio.src /mbond/mbback.src /mc/mcio.src New/Modified Documentation: doc/tpcontrol.doc, dynamc.doc New Testcase: test/c31test/swm4_liquid.inp .............................................................................. 2.2.4 Ambiguously selectable NOE restraints It is often useful to apply a distance restraint between several groups of atoms, not knowing in advance which pair of atoms must be explicitly selected. For example, this happends often when restraints are implemented in a symmetric dimer and one does not know whether a distance restraint is intra or inter molecular. With the key word MINDIST it is now possible to apply such restraint in the NOE module. Files Modified: source/fcm/noe.fcm /energy/energy.src /misc/noe.src /pert/epert.src Modified Documentation: doc/cons.doc .............................................................................. 2.2.5 The RMSD constraint bugfix The rmsd energy constraint as called on each node with MPI. The call to ECNST3 in ecnstr.src is now protected with a "IF(MYNOD.EQ.0)" to be called only once. This is a quick fix for now. A better parallelization of the CONS RMSD could be done in the near future. The passing variables in the call to ECNST3 are corrected. Files involved: source/energy/ecnstr.src ------------------------------------------------------------------------------ 2.3 Developments at University of Vienna Name : Tibor Rudas, Martin Leitgeb, Stefan Boresch, and Othmar Steinhauser Email Address : tibi@mdy.univie.ac.at, martin@mdy.univie.ac.at, and stefan@mdy.univie.ac.at Institution : Molecular Dynamics and Biomolecular Simulation group at the Department of Theoretical Chemistry and Molecular Structural Biology, University of Vienna Date : June 13, 2003 2.3.1 Solvent shell decomposition SHELL decomposes (selected) solvent residues (atoms) into shells based on the distances from (a selected set of) solute atoms. The distance calculation is done by a cubing algorithm (adapted from images/nbndgcm.src by M. Crowley). [The functionality could be duplicated by the trajectory reader and carefully chosen SELEct statements, but the present code is orders of magnitude faster] SHELL can use crystal/images to handle correctly coordinates/trajectories which were generated using periodic boundary conditions with all boxtypes. Lists of atoms in the individual shells, as well as those in the bulk are generated for use in calculations by other modules (e.g. CORREL). Two series using the shell decomposition - SDIP and SATM - are available in CORREL if the SHELL keyword is present in pref.dat (described along with two other new timeseries in CORREL, which were submitted separately). All relevant code is encapsulated by a new keyword - SHELL - so this feature can be safely deactivated. Files Added: source/correl/shell.src, shlsel.src /fcm/shell.fcm Files Modified: source/charmm/charmm_main.src, iniall.src New Documentation: doc/shell.doc New Testcase: test/c31test/shell.inp .............................................................................. 2.3.2 New CORREL time series Correl has been augmented by the following four time series DIPO - dipole moment of (statically) selected atoms (e.g., the dipole moment of a protein or of all waters) SDIP - dipole moment of a water (solvent) shell. (Requires the SHELL module submitted separately. Needs the keyword SHELL to be set in pref.dat) SATM - a true/false (0/1) timeseries if a given atom is in a certain shell (also only available if SHELL is present in pref.dat) VECM - an image-aware analogue to VECT XYZ. Combined with an appropriately chosen CUTIM (see doc.) the vector between the two atoms always represents the minimum image distance. Since two of the series (SDIP and VECM) can/must use images, ANACOR (filling the series from a trajectory) was modified to update images after each frame is read). SDIP and SATM are encapsulated by ##IF SHELL -- ##ENDIF blocks since they can only be used together with the SHELL module. Detailed description of these series is thus also included into shell.doc. Files Modified: source/correl/correl.src, anacor.src Modified Documentation: doc/correl.doc Modified Testcase: test/c31test/cortst2.inp .............................................................................. 2.3.3 Efficient radial distribution function RDFSOL computes radial distribution functions (e.g. g_oo, g_oh and g_hh), as well as some related functions (e.g. the dipole-dipole correlation function). Some of the functions provided can also be calculated by COOR ANALYSIS (solana). The improvements of RDFSOL are twofold: (i) It uses crystal/image instead of calculating minimum distances directly; thus it supports periodic boundary conditions for all boxtypes. (ii) The distance calculations can be done using a cubing algorithm (adapted from images/nbndgcm.src by M. Crowley). While this has a certain performance penalty for small systems, it drastically speeds up calculations for large systems in some cases compared with COOR ANAL. A user function (USRRDF) as fill-function is implemented to enable easy addition of new functions of interest. The complete code is surrounded by a ##IF RDFSOL -- ##ENDIF block to make its compilation optional. Files Added: source/correl/rdfsol.src, rdfsl2.src, shlsel.src Files Modified: source/charmm/charmm_main.src, iniall.src New Documentation: doc/rdfsol.doc New Testcases: test/c31test/rdfsol.inp, rdfsol2.inp .............................................................................. 2.3.4 MMFP GEO restraints in alchemical PERT simulations Presently, the MMFP energy terms are supported by the PERT module only as an invariable energy term, and cannot be part of the alchemical transformation itself. The proposed modification makes MFF-potentials modifiable in PERT free energy simulations. To avoid the overhead of computing GEO restraints twice in situations where they are not part of the alchemical mutation, the new code is only activated if the MMFP keyword is given in the initial PERT statement (see documentation for details). Following PERT MMFP all variables and arrays needed for the specification of the MFF-potentials are duplicated in pert.src. The GEO energy routines are now called twice, once in the lambda = 0 section and once in the lambda = 1 section. Again, existing PERT scripts using GEO restraints as constant energy terms behave as before and should not experience any performance penalty! Files Modified: source/pert/pert.src, epert.src /fcm/pert.fcm /charmm/iniall.src Modified Documentation: doc/pert.doc New Testcase: test/c31test/pert-mmfp.inp .............................................................................. 2.3.5 Exponential flat bottom "Saxon-Wood" potential We have implemented an additional restraint for MMFP based on the Saxon-Wood potential (developed originally in nuclear physics). This potential consists of an almost flat region (of variable width) about the reference value (here the system is practically undisturbed), which is confined by exponentially rising walls, whose height can also be controlled. Thus, the system is restricted very effectively to a predefined region of the phase space. (The potential is similar to a flat-bottom harmonic well, but it is completely smooth and has walls of finite height. The restraining potential was implemented in the MMFP module (see the CHARMM code in mmfp.src) by using existing data structures. Aside from the new keyword SAWO, which must be added in the input script after the GEO command, no additional variables had to be introduced. All existing parameters, which determine the shape, distance and force constant of the restraints, have the usual meaning for the new potential. This type of potential should facilitate applications like binding free energy difference calculations. Files Modified: source/misc/mmfp.src Modified Documentation: doc/mmfp.doc New Testcase: test/c31test/sawo.inp ------------------------------------------------------------------------------ 2.4 UC-Berkeley Developments Name : Aaron R. Dinner Email Address : ardinner@uclink.berkeley.edu Institution : University of California, Berkeley Date : July 3, 2003 2.4.1 Grand canonical Monte Carlo A relatively rudimentary grand canonical Monte Carlo (GCMC) scheme was implemented through the SCALAR, MOVE and MC commands. Note that this code is considered developmental and thus subject to change without backward compatibility. Grand canonical atoms are designated as active through the GCMCon array, which can be manipulated with the SCALAR command. A value of 1 indicates that an atom is active and a value of 0 indicates that an atom is inactive. It is suggested that there be roughly twice as many grand canonical molecules as anticipated will be active on average to accomodate fluctuations. Atoms that block grand canonical insertions in the cavity-based schemes described below are also initialized through a SCALAR array, GCBLocker. A value of 1 indicates that an atom is a blocker, and a value value of 0 indicates that it is not. Time can be saved by excluding hydrogens. Grand canonical moves are declared through MOVE ADD MVTP GCMC. Translation and rotation moves of these molecules should be linked to the GCMC move group to avoid wasting time moving inactive atoms. Two algorithms for facilitating insertions at high density/confinement can be used. The cavity bias method [M. Mezei, Mol. Phys. 40, 901 (1980)] generates a set of candidate insertion positions at each GCMC step randomly, determines the ratio P_N = cavity sites/total sites, and picks a cavity site for insertion. The acceptance probability is adjusted based on the average of P_N to satisfy detailed balance. Simple cavity bias is used if NGCTry is greater than 0. Grid insertion is an improvement of the random cavity bias [M. Mezei, Mol. Phys. 61, 565, (1987)], and works by maintaining a dynamic grid variable within the GCMC volume specifying the cavity sites. It is more memory intensive and slower at lower density, but generally more efficient at higher density and confined geometry such as within the binding pocket of a protein. Grid-based cavity bias is used if RGRId is greater than 0. As indicated above, the insertion volume can be either spherical or rectangular. Current limitations (other than those normally associated with MC): 1. Only atom-based non-bonded lists can be used. 2. The trajectory saved contains all of the grand canonical molecules. The inactive coordinates are set to -9999.9d0 before being written. When using the trajectory file, read the trajectory and then delete the inactive molecules: "DELEte atoms select .byres. prop X .lt. -999.0 end". 3. In grid-based simulations, the grid is updated during non-GC moves using the assumption that all moves translate and rotate species of size comparable to the grand canonical molecules. If you are using grid-based GCMC, with a macromolecular solute, don't allow the solute atoms to move during the GCMC. Files Added: source/fcm/gcmc.fcm /mc/mvgcmc.src /misc/distrib.src Files Modified: source/fcm/mc.fcm /charmm/iniall.src /manip/scalar.src /mc/mc.src, mcener.src, mcimge.src, mcio.src, movead.src, /moveln.src, mvrtrn.src /nbonds/enbfast.src /util/selcta.src .............................................................................. 2.4.2 Momentum-enhanced Hybrid Monte Carlo The method is described in Andricioeai, I., Dinner, A. R. and Karplus, M. (2003) Self-guided enhanced sampling methods for thermodynamic averages. J. Chem. Phys. 118, 1074-1084. The code is encapsulated by the MEHMC keyword in pref.dat. Files Added: source/fcm/mehmc.fcm Files Modified: source/dynamc/dynamc.src /mc/movead.src, mvhmc.src, mc.src .............................................................................. 2.4.3 Transition Path Sampling The Transition Path Sampling (TPS) methods introduced by Chandler and co-workers to sample rare events (see tps.doc) are implemented as extensions of the RXNCoor, DYNAmics, and USER commands in CHARMM. The TPS keyword must be included in pref.dat for the code to be compiled. The code uses the DYNAMC integrator (either leapforg Verlet or Langevin). Files Added: source/dynamc/tps.src Files Modified: source/dynamc/dcntrl.src, dynamc.src /rxncor/rxndef.src, rxnene.src /fcm/rxncom.fcm New Documentation: doc/tps.doc New Testcase: test/c31test/tps_leap.inp ------------------------------------------------------------------------------ 2.5 TReK (version 2.10), a new implementation of CPR Name : Stefan Fischer Email Address : stefan.fischer@iwr.uni-heidelberg.de Date : July 5, 2003 The program TReK (Trajectory REfinement & Kinematics) is a new implementation of a collection of tools for finding the minimum-energy path that joins two known structures representing the reactant and the product states of a reaction. It has been interfaced with CHARMM. The functionality of TReK encompasses and goes beyond that of the old TRAVel module. An effort was made to reproduce the "look & feel" of TRAVEL, whose input-scripts should work with TReK. The main tool is the CPR (Conjugate Peak Refinement) algorithm, which refines an initial guess of the path and finds all the saddle-points. Other tools include the SCM (Synchronous Chain Minimization) algorithm and the SDP (Steepest Descent Path) method, which both allow to smooth a path whose saddle-points are already known. The purpose of the new design of TReK (and main improvement relative to TRAVel) is : - to allow the refinement of long transition paths with hundreds of saddle-points in very large molecules with several thousand atoms. - improved efficiency and robustness for use with quantum potentials. Some of the new features are : 1) Much more user-friendly documentation (./doc/trek.doc). 2) TReK is now parallelized. It scales as well as MD. It was tested under Linux LAM-MPI and MPICH libraries, and under SGI's system MPI libraries. 3) The numerical robustness of the various algorithms has been much improved. 4) An important improvement is the way CPR searches a path for energy-maxima. In TRAVel, a constant step-size was used to scan the whole path and was reduced until it was small enough to adequately probe the most complex region of the energy surface underlying the path. This was inefficient, since the smallest step was used everywhere along the path, resulting in more energy-calls than necessary. In TReK, the step-size varies along the path and automatically adapts to the topography of the energy surface. This is more efficient and makes CPR more robust by preventing the infinite looping that can occur when the step-size is not small enough. 5) CPR writes a restart-file when a path is saved. When the CPR-calculation is continued, this file serves two purposes : - Flagging those path-points that have already been identified as saddle-points, so that they don't have to be flagged manually each time a new TReK session is started. - Saving CPU-time by storing the energy-profile along the path, avoiding the initial energy-scan of the whole path when continuing a CPR refinement in a new TReK session. 6) When the initial guess for a path is very poor or some of its points are very strained, some atoms can experience excessive forces, so that they would get displaced too much during the optimization of the path-points. This is prevented by limiting the rate of displacement to some maximum tolerated value. 7) The energy surface can be expressed in terms of the mass-weighted coordinates, which allows to find the so-called "intrinsic" path. 8) The tool for removing points from a path ("TRAJ DECRease") now can preserve saddle-points and local minima along the path and does not introduce energy peaks. This allows to delete superfluous path-points, saving CPU-time when smoothing the path with SCM. Files Added: source/fcm/trek1.fcm, trek2.fcm Files Removed: doc/travel.doc Files Modified: source/fcm/travel.fcm /rxncor/adiab.src, travel.src, travel2.src /charmm/charmm_main.src New/Modified Documentation: doc/trek.doc, parallel.doc ------------------------------------------------------------------------------ 2.6 Developments at University of Kansas Name : Krzysztof Kuczera Email Address : kkuczera@ku.edu Institution : University of Kansas Date : June 11, 2003 2.6.1 MOLVIB module bug fix and upgrade (c30b1) Eliminated bugs caused by (a) conversion to FORTRAN77 and (b) hidden fixed dimension in routine MATNV. Changed PED to allow systems with above 99 coordinates. Modified array IO to simplify running test cases. Files Modified: source/molvib/molvco.src, molvib.src, gfdiag.src /molvut.src, molvio.src, molvsb.src New test cases: test/c30test/molvt1.inp, molvt2.inp, molvt3.inp, molvt4.inp, /molvt5.inp, molvt6.inp, molvt7.inp .............................................................................. 2.6.2 Conformational free energy thermodynamic integration (c30b1) Test cases and description are expanded for conformational free energy thermodynamic integration (part of TSM module). Files Modified: doc/cfti.doc Files Removed: test/c26test/cftigas.inp, cftmgas.inp New test cases: test/c30test/cftidist.inp, cftiangl.inp, cftidihe.inp, cftmala10.inp, /cftmtst1.inp, cftmtst2.inp, cftmtst3.inp, cftmtst4.inp, /cftmtst5.inp ------------------------------------------------------------------------------ 2.7 Mechanical pulling of two atoms (AFM) Name : Emanuele Paci Email Address : paci@bioc.unizh.ch Institution : University of Zurich Date : June 17, 2003 AFM is an external perturbation designed to pull macromolecules mimicking single molecule experiments (AFM or LOT). There are three possible way of simulating the pulling; all consist in applying a suitable force to two atoms. The force is identical in magnitude for the two atoms, parallel to the two atoms and directed in the direction of increasing distance. The difference in the forces applied concerns their dependence on the time. The three methods currently implemented are: 1) constant force, 2) steered molecular dynamics, 3) biased molecular dynamics (see HQBM.DOC). Files Added: source/fcm/afm.fcm /energy/afm.src Files Modified: source/charmm/charmm_main.src /energy/energy.src, eutil.src /fcm/energy.fcm Added Documentation: doc/afm.doc Added Testcase: test/c31test/afm.inp ------------------------------------------------------------------------------ 2.8 GAMESS-UK Interface Updated (c30b1) Name : Paul Sherwood Email Address : p.sherwood@daresbury.ac.uk Institution : CLRC Daresbury Lab. Date : July 1, 2003 Fix for the special case of Gaussian blur with wmain=0.0 (should implement a point charge but the last version did not). Changes to GAMESS-UK build procedure for use with CHARMM are added to install.com. Integer*8 build is needed for compatibility with some GAMESS-UK builds so this is now supported by install.com (was previously available for GNUALPHA only). This was partially implemented in a previous checking but is complete now. Files Modified: install.com build/UNX/Makefile_alpha source/gukint/gukini.src /machdep/paral3.src /gamint/ddi.src /cadint/cadini.src /flucq/fluqqmmm.src /machdep/space.src /util/util.src, datstr.src ------------------------------------------------------------------------------ 2.9 SUN64 port for running on 64-bit Solaris (c30b1) Name : Bogdan Costescu Email Address : bogdan.costescu@iwr.uni-heidelberg.de Institution : IWR, University of Heidelberg Date : June 6, 2003 Add SUN64 port which allows using 64-bit file access and memory allocations on 64-bit Solaris. A new keyword (SUN64) is introduced to allow code changes with respect to the SUN port; SUN64 is to be used in addition to the SUN keyword. To compile, something like the following line should be used: ./install.com sun64 medium FULL Usage of the "64" install switch is not supported, so the following line is not valid: ./install.com sun medium FULL 64 To avoid breaking output format for other ports, printing of heap-related 64-bit values which do not fit in the existing integer field width is protected by "##IF SUN64". Files Added: build/UNX/Makefile_sun64 Files Modified: install.com source/machdep/machutil.src, space.src /solvation/fft.src /util/util.src ------------------------------------------------------------------------------ 2.10 Developments at NIC-Slovenia Names : Milan Hodoscek Email Address : milan@cmm.ki.si Institution : National Institutes of Chemstry, Ljubljana, Slovenia Date : July 7, 2003 2.10.1 Parallelization of GRAPE interface in CHARMM GRAPE interface in CHARMM was parallelized for more efficient use of a large number of GRAPE boards. Files Modified: doc/parallel.doc, nbonds.doc source/fcm/grape.fcm /nbonds/grape.src .............................................................................. 2.10.2 Support for new version of GAMESS Newest GAMESS version 14 JAN 2003 (R2) is supported in CHARMM Files Modified: source/gamint/gamess/{gamess.src,gmsnbo.src,grd1.src,grd2a.src, guess.src,inputa.src,inputb.src,inputc.src, int1.src,iolib.src,mccas.src,prplib.src, prppop.src,rhfuhf.src,scflib.src, unport.src,zunix.c} .............................................................................. 2.10.3 Support for many non-interacting QM regions within QM/MM system The interface of GAMESS/CHARMM was extended to support the calculations of many QM regions withing QM/MM system which don't interact. The method is extension of the OLAP command in CHARMM to allow for flexible structural alignement of systems with unknown parameters and other docking problems. Files Modified: source/charmm/iniall.src /fcm/gamess.fcm /gukint/gukini.src doc/gamess.doc .............................................................................. 2.10.4 Code cleanups and compiler bugs (c30b1) Some bugfixes for compiler, and minor modifications Files Modified: source/charmm/charmm_main.src /fcm/dimens.fcm (more realistic formula for MAXT&MAXP) /gener/update.src /machdep/socket.c /misc/olap.src /nbonds/nbonda.src, nbonds.src /quantum/addlnat.src /sccdftbint/sccdftbini.src /util/array.src ------------------------------------------------------------------------------ 2.11 Developments at the National Institute of Health Name : Jhih-Wei Chu, Bernhardt Trout and Bernard R. Brooks Email Address : trout@mit.edu, brbrooks@helix.nih.gov Institution : Department of Chemical Engeeniring, MIT; Laboratory of Biophysical Chemistry, NHLBI, NIH Date : July 7, 2003 2.11.1 Nudged elastic bend (NEB) with ABNR support The new Nudged Elastic Bend (NEB) method was developed within RPATH which supports ABNR minimizer. New files: source/fcm/pderiv.fcm, neb.fcm test/c31test/nebdiala.inp, nebdialapmf.inp Files modified: source/charmm/iniall.src /energy/ecnstr.src, energy.src, epath.src /fcm/path.fcm /minmiz/abner.src, egrad1.src, steepd.src /gukint/gukini.src .............................................................................. 2.11.2 Flexible parameter reader The FLEX keyword in READ PARAM command specifies the new flexible parameter format. READ { PARAmeter { CARD [PRINt] [APPEnd] [FLEX] } } { { [FILE] [NBON ] [MMFF] } } This is the same as the standard CHARMM parameter format, but; (1) allows general wildcarding for all terms (2) allows parameter substitution for missing paramters (3) does not require a previously read RTF (no global MASSES list required) (4) allows the definition of paramter equivalence groups. New testcase files: test/c31test/flexpartest.inp /data/flexpartest.prm, flexpartest_h.prm, flexpartest_xh.prm Files modified: doc/io.doc source/charmm/iniall.src /correl/correl.src /dynamc/trnphi.src /fcm/deflts.fcm, param.fcm /gener/genpsf.src, hbonds.src, update.src /image/upimag.src /io/mainio.src, parmio.src /manip/shake.src /misc/hbuild.src /pert/pert.src /quantum/qmset.src .............................................................................. 2.11.3 QUICK command overhauled The quick command has been extended and improved. The following new features; (1) Support for both "old" and "new" syntax (##QUICK_SELECT is removed) (2) Projects positions onto defined AXIS (from COOR AXIS command) (3) Allows the use of comparison coordinates for any specified atom (4) Allows the use of a center of mass/geometry for a group of atoms (5) Allows more ways to select specific atoms. Files modified: source/misc/quicka.src .............................................................................. 2.11.4 Ehancements and Fixes (1) Codes for RMS best fit, CRYSTAL commands, and AXIS commands are improved. Error processing is improved in PERT. A PME bug is fixed in pme.src. Changed the default for Ewald net charge correction (nbonds/nbutil.src). ATOMID call is fixed in manip/cstran.src. Calling argument lists are checked (LISCHECK) and fixed (done on the prerlease version c31a0). (2) prefx.f (prefx.src) is improved to incorporate preprocessing keywords (specified in pref.dat) into the code compiled. (3) SGLD (Self-Guided Langevin Dynamics) feature is separated into dynamc/sglds.src and fcm/sgld.fcm (new documentation is doc/sgld.doc). (4) CMAP code is separated into energy/ecmap.src (5) Hugh memory resiedent arrys are moved into HEAP (polar.src) (6) Source file format is fixed to follow the guideline below. - The last end statement in src files must be in upper case: END. Needed for some legacy compilers. - ##IF construct are grouped with comments in parentheses, e.g., ##IF (pert_code) .... .... ##ELSE (pert_code) .... .... ##ENDIF (pert_code) - ## keyward usage fixed: !##.not. is used instead of !##.NOT. and !##IF is not allowed (though it works). - Unused ## keys are removed (in progress) : OPLS, QUICK_SELECT and others. (7) documentation updated. New Files: doc/sgld.doc, preflx_list.doc source/dynamc/sglds.src /energy/ecmap.src /fcm/sgld.fcm test/c31test/sgld.inp Files removed: source/dynamc/meanf.src /fcm/fbias.fcm Files modified: doc/corman.doc, crystal.doc, developer.doc, dynamc.doc, io.doc, /lonepair.doc, minmiz.doc, miscom.doc, nbonds.doc, perturb.doc, /subst.doc source/adumb/cmds.src, eadumb.src, svdcmp.src /charmm/iniall.src /cheq/fqcom.src, fqener.src, fqminut.src /correl/clustr.src, rdfsl2.src, rdfsol.src, shell.src, solana.src /dynamc/dcntrl.src, dynamc4.src, dynamc.src, dynamcv.src, /dynamln.src, dynamvv2.src, dynamvv.src, dynsub.src, dynio.src, /tpcontrol.src, tps,src, trnphi.src, tmd.src /emap/emapio.c, emapsubs.src /energy/cenerf.src, ebaspas.src, ecnstr.src, epath.src, eintern.src /gbmv.src, printe.src, polar.src, rgy.src, rmsd.src /fcm/afm.src, cff.fcm, cheqdyn.fcm, clcg.fcm, cnst.fcm, /consta.fcm, derivq.fcm, eef1.fcm, emap.fcm, /excl.fcm, exfunc.fcm, flucq.fcm, gamess.fcm, /gbmv.fcm, gbsw.fcm, gcmc.fcm, genborn.fcm, lambda.fcm, /leps.fcm, lobos.fcm, mbmts1.fcm, mbmts.fcm, mbond.fcm, /mbpar.fcm, mbspp.fcm, mc.fcm, mehmc.fcm, mltcanon.fcm, /olap.fcm, path.fcm, pathint.fcm, pert.fcm, rtf.fcm, /rwlamb.fcm, sasa.fcm, sccdftb.fcm, spacod.fcm, spapar.fcm, /traj.fcm, travel.fcm, tsmh.fcm, tsms.fcm, umb.fcm /gamint/ddi.src /gener/drude.src, mkpres.src, modpsf.src, update.src /graphics/graphx.src /gukint/gukini.src /image/cnbndm.src, nbndgcm.src, crystal.src, pbound.src /io/coorio.src, psfres.src, rtfio.src, trajio.src, univio.src /machdep/paral1.src, corman.src /manip/corman2.src, corman3.src, cshake.src, cstran.src, /dynanal.src, hbanal.src, helix.src, rmsdyn.src, /rotlsq.src, scalar.src /mbond/mbback.src, modtrans.src /mc/mcener.src, mcio.src, mcmini.src, moveln.src /minmiz/abner.src, conjug.src, egrad1.src, minmiz.src, /nraph.src, powell.src, steepd.src, tndriv.src /misc/aspener.src, distrib.src, genetic.src, nmr.src, xray.src /mmff/merckio.src /nbonds/enbf2.src, enbfast.src, enbondg.src, enbond.src, /evdwf.src, grape.src, nbutil.src, pme.src, pmeutil.src /pert/block.src, epert.src, pert.src, puic.src, tsms.src /quantum/addlnat.src /rxncor/adiab.src, lupopt.src, rxndef.src, rxnene.src, /travel2.src /solvation/rismio.src /util/clcg.src, imsl.src, prpref.src, selcta.src, string.src /vibran/quasi.src, rbquas.src, vibio.src tool/prefx.f, prefx.src ------------------------------------------------------------------------------ 2.12 HEAP allocation and release status information Name : Youngdo Won Email Address : won@sulu.hanyang.ac.kr Institution : Hanynag University and NIH Date : April 25, 2003 An array is allocated on the heap by the ALLHP function and is released by the FREHP subroutine. Upon allocating/releasing the heap space, it is nice to print the information on the dynamic heap usage. When the heap-print-out option is set (by the TEST HEAP command), (1) the ALLHP function is modified to print the message: ALLHP> allocates words from
for (2) the FREHP subroutine is modified to print the message: FREHP> releases words of at
The calling argument list of ALLHP and FREHP is modified to add , , and , for example, HPNDIS = ALLHP(IREAL8(NDISTA),'cmds.src','UMBAN2','HPNDIS') CALL FREHP(HPNIDS,IREAL8(NDISTA),'cmds.src','UMBAN2','HPNDIS') ALLHP and FREHP of util/util.src are modified to put the information printing when PRHEAP is set. Note that the calling argument list of ALLHP and FREHP is changed. Files Modified: (142 source files) adumb/cmds.src, umb.src charmm/charmm_main.src, iniall.src cheq/fqener.src correl/correl.src, solana.src dimb/dimbutil.src, nmdimb.src dynamc/dcntrl.src, nose.src, tmd.src emap/emapsubs.src energy/dmcons.src, ecnstr.src, ecntrl.src, ediff.src, enefvp.src, /energy.src, eolap.src, epath.src, epull.src, gbmv.src, genborn.src, /intere.src, pathint.src, rgy.src, rmsd.src flucq/flucq.src, fluqqmmm.src gener/genpsf.src, makphi.src, modpsf.src graphics/graphx.src image/crystal.src, eimage.src, images.src, imnbfp.src, nbndgcm.src, /upimag.src, xtlfrq.src io/coorio.src, parmio.src, trajio.src machdep/paral1.src manip/corman2.src, corman3.src, corman.src, cshake.src, cstran.src, /fsshake.src, hbanal.src, intcor.src, rmsdyn.src, scalar.src mbond/mbmts.src mc/mcace.src, mcener.src, mcimge.src, mcio.src, mcmini.src, mc.src, /movead.src, moveio.src, moveln.src, mvcrot.src, mvdihe.src, mvhmc.src, /mvrtrn.src minmiz/abner.src, egrad1.src, nraph.src, powell.src, steepd.src, tndriv.src misc/eef1.src, genetic.src, grid.src, gsbp2.src, gsbp.src, mmfp.src, /nmr.src, olap.src, pbeq2.src, pbeq.src, primsh.src, resdist.src, /sasini.src, ssbp.src, surfac.src, testch.src, zmat.src mmff/efast_mm.src, enbfast_mm.src molvib/molvco.src nbonds/enbond.src, etable.src, evdwf.src, ewaldf2.src, ewaldf.src, /fma.src, nbexcl.src, nbndcc.src, nbndfp.src, nbndgc.src, nbonds.src, /nbutil.src, pme.src, pmeutil.src pert/block.src, epert.src, icfcnf.src, icpert.src, pert.src, puic.src, /tsme.src, tsms.src, wham.src quantum/qmene.src, qmjunc.src, qmnbnd.src, qmset.src rxncor/adiab.src, lupopt.src, path.src, rxndef.src, travel2.src, travel.src shapes/eshape.src, shapes.src solvation/cycles.src, deriv.src, rism.src, solvation.src util/datstr.src, pointer.src, selcta.src, util.src vibran/vibio.src, vibran.src, vibsub.src, vibutil.src ============================================================================== [3] c31a2 Change Log The following are introduced and modified during the development period of August 15, 2003 through February 15, 2004. 3.1 TSRI Developments Name : Sandeep Patel, Jainhan Chen, Scott Brozell, Mike Crowley, Charles L. Brooks, III Email Address : sandeep@scripps.edu, jainhan@scripps.edu, crowley@sripps.edu, brooks@scripps.edu Institution : The Scripps Research Institute Date : December 29, 2003 3.1.1 Charge Equilibration Polarizability Enhancements and Fixes Allocate heap space for wall potential and charge mass arrays if CHEQ requested. Removed unnecessary comments and variables in various routines. Fixed routine GAUSSJ in cheq/fqener.src (declaration of variable PIVINV). Files Modified: source/charmm/charmm_main.src, iniall.src /cheq/fqcom.src, fqener.src /dynamc/dynamc.src, dynamvv.src /energy/energy.src /fcm/cheqdyn.fcm /mbond/mbback.src .............................................................................. 3.1.2 NOE Potential with Soft Asymptote Implementation of a soft-square NOE potential where the square-well function is used for distances within a specified "switching" region (specified by RSWIitch keyword) and outside this region a "soft" asymptote is used: / 0.5*KMIN*(RAVE-RMIN)**2 R<=RMIN E(NOE) = | 0.0 RMIN RMAX+RSWITCH In addition, an option (invoked by SUMR) of using summation instead of averaging for calcuating R was added: R = [ Sum_ij ( Rij**(1/REXP) ) ]**REXP Files Modified: source/adumb/cmds.src, eadumb.src /energy/energy.src /fcm/noe.fcm, pert.fcm, umb.fcm /mc/mcener.src, netic.src /misc/noe.src /pert/epert.src Modified Documentation: doc/cons.doc .............................................................................. 3.1.3 Endian-checking Trajectory Reader Reads trajectories transparently to the user regardless of whether it is big or little endian. Saves having to do dyna format and dyna unformat on large trajectories that come from machines with opposite endianness. NOTE: This automatic file conversion does not yet work properly on IA64 architectures, will be fixed soon. Files Modified: source/dynamc/dynio.src, dynsub.src /fcm/iochan.fcm /machdep/cstuff.c, machio.src /util/title.src New Testcase: test/c31test/endian.inp /data/endian_0.crd, endian_1.crd .............................................................................. 3.1.4 Intel 64 bit Machine Ports Added support for new machine architecture and compilers. The new install.com keys and switches are hpitanium for HP-UX Itaniums altix for SGI Altix series (64 bit) gnu EFC for efc intel 64 bit compiler Roberto Gomperts added mpi8 routines. These convert 8 byte integer arguments for charmm built with I8 into 4 byte integer arguments for mpi calls and are guarded by the pref keyword ALTIX_MPI which is activated by the install.com target altix. This mechanism enables coexistence with other mpi implementations, such as LAM, that do not support MPI_INTEGER8. Modified to build 64 bit on Linux Itaniums using install keyword EFC. Roberto Gomperts identified some mpi_allreduce calls that were not passing the correct message lengths for 64 bit integers. A general solution would be to use MPI_INTEGER8, but that is not supported by all mpi implementations yet. So the message count is doubled. Files Added: build/UNX/Makefile_altix, Makefile_hpitanium Files Modified: install.com build/UNX/Makefile_gnu source/correl/solana.src /energy/gbsw.src /fcm/impnon.fcm /machdep/machio.src, machutil.src, paral1.src, paral2.src /space.src, startup.src /manip/shake.src /nbonds/pmeutil.src .............................................................................. 3.1.5 McIntosh OSX Port Port added for for Macintosh OSX. The install.com target is set to OSX and the compiler switch is either xlf or g77. The pref.dat keywork is OSX. install.com osx [ xlf | g77 ] Files Added: build/UNX/Makefile_osx Files Modified: install.com build/UNX/misc.mk, solvation.mk source/cadint/cadini.src /correl/solana.src /emap/emapio.c /fcm/impnon.fcm /machdep/cstuff.c, machio.src, machutil.src, paral1.src /paral2.src, parset.src, space.src, startup.src /quantum/qmset.src .............................................................................. 3.1.6 Testcase Validation A full set of output for all tests is placed in test/Testoutput for comparison and validation purposes. It is intended to be used with test.com as follows: test.com gnu G20 Testoutput 20 for the c20test tests to be run, where G20 is the test output directory. The result is compared with the output in Testoutput, and the pared-down diffs placed in G20.rpt. A validation awk script is also submitted, comare.awk, which is to be used to determine pass/failure and to report the relevant diffs only. awk -f compare.awk G30.rpt will process G20.rpt and report PASS of FAIL based on relevant diffs only. A more verbose output with the relevant diffs reported is produced with: awk -f compare.awk verbose=1 G20.rpt Some tests were modified to get rid of extraneous stuff. Files Added: test/compare.awk Files Modified: test/c20test/noetest.inp test/c23test/nose1.inp, nose2.inp test/c24test/pbound1.inp test/c26test/pathint.inp, tpnoe.inp test/c27test/testeef1.inp, testeef1imag.inp test/c30test/cq1ab1gbdyn.inp, cq1fsvvacu.inp, cq1fsvwat.inp /cqmeohvacu.inp, cqnmawat.inp, cqtip4pbulk.inp .............................................................................. 3.1.7 Bugfixes Fixes to code for compaq alpha mostly centered on emap code. Some fixes are submitted here and at least the code compiles now, but emap still broken. (1) Added a note that GBMV requires GENBORN prefx name for compilation. (2) Fixed FASTEW known incompatibility for CHEQ. (3) Fixed timing bug in dynamics (4) arg list missing vars in epath.src (5) Fixed problem with integer*4 and integer*1 declarations in GBMV (6) Bugfix of problem printing dihe constraints. (7) GBSW source code moved from misc/ to energy/ (8) adumb/svdcmp.src modernization of do loops to accomodate f90 compiler (9) SHELL code logical variable fix in anacor.src, correl.src, shell.src (10) Timing bug in dynamics fixed. (11) energy.src: Changes to fix redundant call to lonepair force remapping. (12) gbmv.src: Removed integer*4/*1 declarations for correct performance on ia64 and xlf machines. Modified to improve readability and computational performance. (13) gbsw timers and constants fixed. (14) mcener.src and movead.src: PATHINT temporary fixes (15) mcmini.src: BROKEN protection from steepd/neb disfunctional code (16) abner.src: NEB problem moved inside if block (17) nbexcl.src: fix for lrvdw when it is not specified in the nbond parameters File moved: source/energy/gbsw.src from source/misc Files Modified: build/UNX/Makefile_alpha, energy.mk, misc.mk source/adumb/svdcmp.src correl/anacor.src, correl.src, shell.src dynamc/dcntrl.src, dynamc.src emap/emapio.c, emapsubs.src, energy/energy.src, epath.src, gbmv.src, gbsw.src fcm/new_time.fcm, shell.fcm io/parmio.src machdep/cstuff.c, socket.c manip/cstran.src mbond/mbback.src mc/mcener.src, mcmini.src, movead.src, mvcrot.src, mvhmc.src, mvrtrn.src minmiz/abner.src, steepd.src nbonds/ewald.src, nbexcl.src, nbutil.src util/new_timer.src Modified Documentation: doc/gbmv.doc, cheq.doc ------------------------------------------------------------------------------ 3.2 Developments at the National Institute of Health Name : Bernard R. Brooks Email Address : brbrooks@helix.nih.gov Institution : Laboratory of Biophysical Chemistry, NHLBI, NIH Date : December 31, 2003 3.2.1 Holonomic constraints The modification intends creation of a central point for all holonomic constraints to ensure compatibility and consistency. This provides a template for adding additive constraints. File Added: source/manip/holonom.src Files Modified: source/dynamc/dynamc.src, dynamc4.src, dynamcv.src, dynamln.src /dynamvv.src, dynamvv2.src, mts.src, prssre.src, tps.src /energy/energy.src, eutil.src /manip/corman.src, shake.src /minmiz/egrad1.src /vibran/vibio.src, vibsub.src, vibutil.src .............................................................................. 3.2.2 CODES Extension Extension to CODES to allow it to be called with a subset of atoms. This allows non-CHARMM force fields to use varrious commands, e.g., IC PARAM, that require a parameter lookup for particular atoms. Files Modified: source/correl/correl.src /gener/genpsf.src, update.src /manip/intcor.src, intcor2.src, shake.src /misc/hbuild.src /pert/pert.src /quantum/qmset.src .............................................................................. 3.2.3 Callable Parser Main program split. Main parser is now a callable routine and allows a CHARMM command to invoke another command. File Modified: source/charmm/charmm_main.src .............................................................................. 3.2.4 VECTOR Codes Removed Old FAST codes (PARVECT, CRAYVEC, CRAY, VECTOR) are removed. The FASTer command allows three stage options, OFF, ON and DEFAult. The default is to use fast scalar routines where available. Files Removed: source/energy/cenerf.src, enefst2.src, enefst2q.src, enefvect.src, enefvp.src /fcm/cray.fcm /image/cnbndm.src, imnbf2p.src, imnbfp.src /mmff/enbfast_mm.src /manip/cshake.src /nbonds/cenbf.src, cnbnd.src, enbf2.src, evdwf.src, ewaldf2.src /nbonds/nbndf2p.src, nbndfp.src Files Modified: source/cff/enbfast_cff.src /charmm/iniall.src, miscom.src /energy/anal.src, ecntrl.src, enefscal.src, epath.src /fcm/dimens.fcm, fast.fcm /fcm/ /mc/mcener.src, mcimge.src /mc/ /mmff/efast_mm.src /nbonds/enbfast.src, enbond.src, ewaldf.src, nbonds.src /pert/epert.src .............................................................................. 3.2.5 Bugfixes and Clean-up General cleanup, improving ## control... Files Modified: source/dynamc/dcntrl.src, dynio.src, lonepair.src, tpcontrol.src, trnphi.src /fcm/fstshk.fcm, param.fcm, timer.fcm, tmd.fcm /io/coorio.src, parmio.src /manip/rotlsq.src /mc/mcio.src /misc/quicka.src /mmff/enbscalar_mm.src /nbonds/ewald.src ------------------------------------------------------------------------------ 3.3 Developments at NIC-Slovenia Names : Milan Hodoscek Email Address : milan@cmm.ki.si Institution : National Institutes of Chemstry, Ljubljana, Slovenia Date : July 7, 2003 3.3.1 Parallelization of TMD code Parallelization of TMD method completed Files Modified: doc/parallel.doc source/dynamc/tmd.src /energy/rmsd.src, ecnstr.src /manip/rotlsq.src .............................................................................. 3.3.2 New global variable MDSTEP Introduction of global variable for number of steps in the dynamics. Before this was only for GRAPE module, now it is global. Files Modified: source/fcm/contrl.fcm, grape.fcm /dynamc/dynamc.src /nbonds/grape.src .............................................................................. 3.3.3 Trajectory files for QM charges from dynamics calculations Writing of QM charges from GAMESS in the trajectory format. Since GAMESS calculates 3 types of charges they are put into the trajectory file in the following order: X - Mulliken; Y - Lowdin; Z - ESP charges. Files Modified: source/fcm/gamess.fcm, reawri.fcm /charmm/iniall.src /dynamc/dcntrl.src, dynamc.src /gukint/gukini.src .............................................................................. 3.3.4. Support for new version of GAMESS Newest GAMESS version 3 JUL 2003 (R1) is supported in CHARMM and some fixes for the QM/MM interface Files Modified: source/fcm/gamess.fcm /gukint/gukini.src source/gamint/gamess/gamess.src,gmsnbo.src,grd1.src,grd2a.src, /guess.src,inputa.src,inputb.src,inputc.src, /int1.src,iolib.src,mccas.src,prplib.src, /prppop.src,rhfuhf.src,scflib.src, /unport.src,zunix.c ------------------------------------------------------------------------------ 3.4 Generalized Born-Surface Area Model with Implicit Membrane (GBSA/IM) Name : Velin Spassov, Lisa Yan, Sandor Szalma, Piotr Cieplak, Krzysztof Olszewski Email Address : kato@accelrys.com Institution : Accelrys Date : 12/18/03 GBSA/IM permits calculations of solvation energy and forces in presence of Implicit Membrane as it is described in Spassov,V.Z, Yan, L and Szalma S. J. Phys. Chem. B, 106,8726-8738 (2002). The calculations are based on the simultaneous use of two CHARMM modules: GBIM and ASPENRMB. GBIM module includes the effect of Implicit Membrane (IM) in the calculations of the electrostatic contribution to solvation energy in the frames of Generalized Born approach (see gbim.doc). GBIM module is extended version of GENBORN module of B. Dominy and C.L. Brooks,III (genborn.doc) where the membrane is added to molecular model as a planar dielectric slab having the same dielectric constant as inside the molecule. In addition to standard Gborn parameters, GBIM has optional parameters to control the intramolecular dielectric constant, membrane thickness as well as the position and orientation of membrane relative to coordinate system. ASPENRMB is modification of ASPENR (aspenr.doc) by including Implicit Membrane (IM) in Surface Area (SA) calculations. In GBSA/IM approach ASPENRMB is used to calculate the non-polar contribution to solvation energy. ASPENERMB can be also as a separate program to calculate solvation energy and forces based on the solvent accessible surface area of each atom using Atomic Solvation Parameters (ASP), as proposed by Wesson and Eisenberg. In addition to standard ASP parameters, ASPENRMB uses optional parameters to control the thickness as well as the position and orientation of membrane relative to coordinate system. ASPENRMB is activated by reading a surf file, as described in doc/aspenrmb.doc. keywords: GBIM ASPMEMB Remark: GENBORN and ASPENER must be present in pref.dat as well. Files Added: source/energy/gbim.src /fcm/gbim.fcm, surfmemb.fcm /misc/aspenermb.src Files Modified: install.com source/charmm/charmm_main.src, iniall.src /energy/energy.src, intere.src /io/mainio.src New Documentation: doc/gbim.doc, aspenrmb.doc New Testcase: test/c31test/gbsaim.inp ------------------------------------------------------------------------------ 3.5 Screened Coulomb Potentials Implicit Solvent Model (SCPISM) Names : Sergio A. Hassan and Ernest L. Mehler Email Address : mago@helix.nih.gov Institution : CMM/DCB/CIT, National Institutes of Health (NIH); WMC, Cornell University; Mount Sinai School of Medicine (NYU) Date : December 19, 2003 The SCPISM is an implicit solvent model (ISM) for macromolecules. Parameterization of the model was carried out only for proteins, and for use with the all-atom PAR22 forcefield. The current version incorporates only a continuum approach for electrostatics and optionally a simple term that accounts for hydrophobic effects (see documentation in scpism.doc for details). The model is based on screened Coulomb potentials (SCP). The current implementation is the initial step in a long-term project aimed at developing a general implicit solvent model (ISM) for macromoleculs, i.e., a continuum approach to describe both electrostatic and non-electrostatic effects, to be used in large-scale molecular mechanics simulations. The current implementation is suitable only for dynamics simulation, energy evaluation and minimization of peptides and proteins (any CHARMM options are supported) Files Added: doc/scpism.doc source/fcm/scpism.fcm /misc/scpism.src test/c31test/scpismtest.inp /data/scpism.inp Files Modified: source/charmm/charmm_main.src /nbonds/enbond.src ------------------------------------------------------------------------------ 3.6 The University of Maryland Developments Name : Victor Anisimov and Alexander D. MacKerell, Jr. Email Address : victor@outerbanks.umaryland.edu Institution : The University of Maryland School of Pharmacy Date : December 29, 2003 3.6.1 Drude Oscillator Code Fixes With the Drude polarizable model situations arise where atoms have more than 8 bonds. A switch has been added that increases IATMXB to 20 when Drude particles are present in the system versus the default value 8. Declaration of KDRUDE variable is moved from drude.src to the PSFR common block from which it can be shared by other routines. Files Modified: source/fcm/psf.fcm /gener/drude.src /nbonds/nbexcl.src .............................................................................. 3.6.2 Gaussian Input Generation Gaussian input generation was changed to be compliant with Gaussian98 and Gaussian03 requirements. The three last rows of comments in the input file created by CHARMM were removed since modern Gaussian misinterprets them as ONIOM instructions. The old Gaussian input generator allows printing coordinates of those atoms only which have Z-matrix records. Other atoms will be omitted thereby limiting the utility of the created file. The new generator writes mixed Cartesian-internal coordinate file where all atoms present in the system will be printed. This mode is activated by IC GAUSS ZMIX keyword. The mixed coordinate format is particularily useful for partial QM geometry optimization of two rigid interacting molecules, e.g. a model compound and a water molecule. Such cases are common in CHARMM parameter optimization procedures where CHARMM optimized interaction energies and geometries of two interacting molecules should be verified against QM results. The first molecule is defined in the Gaussian input file by Cartesian coordinates and second one is represented by internal coordinates in an orientation relative to the first molecule. Files Modified: source/manip/wrgaus.src, intcor.src doc/intcor.doc .............................................................................. 3.6.3 Fixed Origin for DIPOLE Calculation Standard dipole calculation by CHARMM moves the dipole origin of ions into geometric center of molecule. This becomes problematic when comparing the CHARMM dipole moment with Gaussian QM data, since Gaussian does not move the dipole origin. The OXYZ keyword is added to prevent the origin move in CHARMM. The format of the command is COOR DIPOle OXYZ. Files Modified: source/correl/anacor.src, solana.src /manip/corman.src doc/corman.doc .............................................................................. 3.6.2 CMAP bugfix With all trans phi, psi geometry CMAP energies are wrong. This is due to a machine precision problems. offset values to avoid precision problem. File fixed: soruce/energy/ecmap.src ------------------------------------------------------------------------------ 3.7 Karolinska Institute Fixes and Enhancements Name : Lennart Nilsson Email Address : Lennart.Nilsson@biosci.ki.se Institution : Karolinska Institute Date : January 3, 2004 1. Added possibility to specify a "MODEL" to be read/written using READ/WRITe COOR PDB. Files modified: io/coorio.src doc/io.doc 2. Modified CORRel timeseries INERtia to allow user to extract the three main moments of inertia of selected atoms. Files modified: correl/correl.src doc/correl.doc 3. Added conversion of cartesian vector timeseries to spherical coordinates in CORRel MANTime. Files modified: correl/mantim.src doc/correl.doc 4. Modification to allow CORRel EDIT to modify VECCOD of individual timeseries (previous was to change VECCOD for all components of a vector timeseries) Files modified: correl/correl.src 5. Several modifications to NMR-module. Allow compact output of relaxation parameters to file; write files that can be used as input for ModelFree program; estimate entropy and effective correlation times of eg backbone NH vectors; allow CSA contribution to relaxation rates; allow user to turn off analytic calculation of overall rotation diffusion. Files modified: source/misc/nmr.src doc/nmr.doc 6. Bugfix in calculation of solvent-site radial distribution function. Files modified: correl/solana.src 7. Bugfix in output of integration timestep from GTICNT. Files modified: dynamc/dynio.src ------------------------------------------------------------------------------ 3.8 Harvard Developments Name : Robert Petrella Email Address : petrella@tammy.harvard.edu Institution : Harvard University Date : January 3, 2004 (1) Clustering for BYCC set up to be done by default just after the exclusions are first made (after first call to UPDATE). The NO_BYCC preprocessor "if" statements in the cluster-related algorithms were removed in order to make this a default process. Hence, the MKCLusters command is no longer necessary to use BYCC. These changes also allow "active atom" selections for non-bonded interactions to be coded easily into routines anywhere in charmm. Files modified: source/nbonds/nbutil.src, nbexcl.src, nbonds.src, nbndcc.src /charmm/iniall.src /fcm/actclus.fcm, exclar.fcm (2) ATOMON array in eef1 was replaced by a global active atom array, ACAFLG, which is either 1 or 0 depending on whether the atom is active or not. This fixes a bug involving CONS FIX, eef1, and isolated atoms. Files modified: source/misc/eef1.src /energy/energy.src, intere.src /mc/mcener.src (3) Compiler optimization macros in build/UNX/nbonds.mk were changed from FC0 to FC2 in the following cases: enbondg.f, nbndcc.f, nbndgc.f and nbondg.f. This increases the speed of both the default (BYGR) and BYCC list-generation routines, without introducing any obvious compilation issues. (4) Minor additional modifications to nbutil.src for printing out correct non-bonded options. File modified: source/nbonds/nbutil.src 5) BYCC now handles atom clusters that are larger than the non-bonded cutoff distance. source/nbonds/nbndcc.src ------------------------------------------------------------------------------ 3.9 University of Wisconsin Developments Name : Qiang Cui and Guohui Li Email Address : cui@chem.wisc.edu, gli@chem.wisc.edu Institution : University of Wisconsin, Madison Date : December 23, 2003 3.9.1 Block Normal Mode method and Gaussian (Anisotropic) Network Model The BNM approach projects the full atomic hessian into a subspace spanned by the eigenvectors of blocks, which reduces the size of the eigenvalue problem dramatically. Each block can be defined by the user in very flexible manners as an amino acid or a secondary structural element. Currently only the T/R vectors of the blocks are included as basis vectors. The main code is vibblock.src, which constructs the projected hessian in a direct manner, which is essential for applications to large systems (e.g., ribosome). Vibblock.src requires input from vibran.src, and it (through SUBROUTINE BLOCKHESSIAN) calls the subroutines that handle 2nd derivative calculations. In the current version, only BOND, ANGLE, UREY-BOND, DIHE, ImproperDIHE, Electrostatic, VDW and EEF1 terms are supported. Three new variables (QBSECD,HSBLC,MAXD) have been added to the list of arguments of these subroutines. All source files (see the modified files list) that call these energy subroutines have been changed; i.e., ".false.,0,0" was appended. Both serial and parallel versions are supported. The parallel version has only been tested on linux clusters with the "gnu M mpich" options. For the diagonalization of large sparse matrices, the BNM code uses the (P)ARPACK package, which can be downloaded for free from the PARPACK website. The Gaussian (Anisotropic) Network Model(GANM) has also been implemented, and only the serial version is available now. The calculation needs two files made by the user. Files Added: source/vibran/vibblock.src /fcm/vibblock.fcm Files Modified: doc/vibran.doc source/correl/anacor.src /energy/energy.src, eintern.src, intere.src, ediff.src /ecntrl.src, ecnstr.src /image/eimage.src /mbond/mbdiag.src /mc/mcener.src /misc/genetic.src, hbuild.src, eef1.src /nbonds/nbonds.src, nbondg.src, nbonda.src, enbondg.src, /enbonda.src, enbond.src /pert/icfcnf.src, epert.src, icpert.src, tsme.src /quantum/qmene.src /vibran/vibran.src New Testcases: test/c31test/bnmtest.inp, ganmtest.inp .............................................................................. 3.9.2 SCC-DFTB/MM Free Energy Calculation A Dual-Topology-Single-Coordinate (DTSC) approach has been implemented for SCC-DFTB/MM free energy perturbation calculations. With DTSC, which is very similar to PERT, free energy perturbations can be carried out with SCC-DFTB/MM for chemical species with very similar chemical strctures but different electronic distributions (e.g., redox processes). With minor modifications, the code can now be used to evaluate pKa's. It MUST be used with "FAST OFF" in the CHARMM script because only generic atom-atom codes have been modified so far. Periodic boundary condition and cutoff for SCC-DFTB/MM electrostatics have also been put in to facilitate solution studies; only "Force-based" or "Energy-based" SHIFTs have been implemented. Files Added: source/fcm/blockscc.fcm Files Modified: doc/sccdftb.doc source/energy/energy.src, eintern.src /dynamc/dynio.src, dcntrl.src, dynamc.src, dynamc4.src /dynamcv.src, dynamvv.src /fcm/sccdftb.fcm, gamess.fcm /gener/update.src /image/upimag.src /io/coorio.src /mbond/mbback.src /mc/mcio.src /nbonds/enbonda.src, heurist.src /pert/block.src /sccdftbint/sccdftbini.src /sccdftbsrc/dylcao.f, eglcao.f, externalchgrad.f, /externalshift.f, gettab.f, mulliken.f, shift.f [Note: testcases will be submitted by June 30, 2004.] ------------------------------------------------------------------------------ 3.10 Force Field Developments Name : Alex MacKerell Email Address : alex@outerbanks.umaryland.edu Institution : Univ. of Maryland School of Pharmacy Date : January 19, 2004 3.10.1 Restructurinf RTF/PARAM Files In version 31 a major restructuring of the topology and parameter files was undertaken. This was performed to create a more modular approach to the files, thereby avoiding the problem of files becoming increasingly large. The restructuring was done such that the primary topology and parameter files for biomolecules can be used as they were previously. However, the topology and parameter information for the majority of model compounds used in the parameter development, additional molecules, including coenzymes, and patches parametrized to be compatible with the CHARMM force fields have been moved to toppar stream files. These toppar stream files have to be streamed in a CHARMM input script typically following reading of the parent topology and parameter files. They include both the topology and parameter information for the selected molecules, using the "read rtf card append" and "read param card append" commands to append the additional information to the topology and parameter lists. As of version c31 it is still necessary to maintain all the MASS atom lists in the parent topology and parameter files. Modified Documentation: doc/energy.doc, io.doc, parmfile.doc, rtop.doc .............................................................................. 3.10.2 CMAP Parameter Addition These are identical to the charmm22 force field except for the addition of CMAP. Work is in progress on additional corrections to the charmm22 force field that will be released as all31 or a later version. New protein topology and parameter files added: toppar/top_all22_prot_cmap.inp, par_all22_prot_cmap.inp .............................................................................. 3.10.3 Non-CHARMM Parameters topology and parameter files for BMS, AMBER and OPLS force fields. See 00readme in the toppar/non_charmm subdirectory. toppar/non_charmm/par_amber_98.inp, par_amber_cornell.inp /par_bms_dec03.inp, par_opls_aa.inp /par_opls_ligand.inp, top_amber_cornell.inp /top_bms_dec03.inp, top_opls_aa.inp ============================================================================== [4] c31b1 Change Log The following are introduced and modified during the development period of February 15, 2004 through August 15, 2004. The developments are based on CHARMM c31a2. As declared at the CHARMM Meeting (Harvard University, July 9-10, 2004), the following features are still under development and not included in the c31 release versions: Grand Canonical Monte Carlo (GCMC) GCMC Block Normal Mode method VIBBLOCK Gaussian (Anisotropic) Network Model VIBBLOCK SCC-DFTB/MM Free Energy Calculation SCCDFTB The SPASIBA Force Field SPASIBA Endian Testing TESTENDIAN ------------------------------------------------------------------------------ 4.1 TSRI Developments Developers : Scott Brozell, Mike Crowley, Charles L. Brooks, III Roberto GOmpers (SGI) Email Address : sbrozell@scripps.edu, crowley@sripps.edu, brooks@scripps.edu Institution : The Scripps Research Institute Date : July 4, 2004 4.1.1 4-Byte Integer Binary IO 64 bit integers are default for many architectures now. Trajectories and other binary files written by charmm are unreadable on other machines when the binary files contain 64 bit integers. This mod makes all binary io restricted to 32 bit integers only Files Modified: install.com source/correl/solana.src /dimb/dimbsub.src /dynamc/dynio.src /dynsub.src /fcm/iochan.fcm /deflts.fcm /gener/hbonds.src /image/xtlfrq.src /io/coorio.src /parmio.src /psfres.src /machdep/cstuff.c /machio.src /manip/corman2.src /intcor.src /molvib/molvio.src /nbonds/etable.src /util/datstr.src /title.src /util.src /vibran/rbquas.src /redbas.src vibio.src Modified Testcase: c31test/endian.inp .............................................................................. 4.1.2 Performance Enhancement of ENBFS8 Loop rearrangement and temporary arrays for locality to speed up fast nonbond routine. Files Added: source/fcm/enbfs8p.fcm source/nbonds/enbfs8p.src Files Modified: install.com source/nbonds/enbfast.src /util/new_timer.src .............................................................................. 4.1.3 64 bit Parallel and Serial IBMAIX Codes (Mike, Could you put a few words here?) Files Added: build/UNX/Makefile_ibmaix /Makefile_ibmaix64 /Makefile_ibmaixmp64 Files Modified: install.com source/correl/solana.src /machdep/cstuff.c /paral1.src /paral2.src /space.src /manip/shake.src /nbonds/pmeutil.src ------------------------------------------------------------------------------ 4.2 Developments at the National Institute of Health Developers : H. Lee Woodcock and Bernard R. Brooks Email Address : hlwood@nih.gov, brb@nih.gov Institution : Laboratory of Biophysical Chemistry, NHLBI, NIH Date : July 1, 2004 4.2.1 Autogenerate for Patching Option An autogenerate feature has been added to the RTF reader and the PATCh command which allow the autogeneration of angles and dihedrals. The PATCH command syntax: PATCh misc-patch-spec residue-list-spec misc-patch-spec ::= [SORT] [SETUp] [WARN] [ ANGLe ] [ DIHEdrals ] [ NOANgle ] [ NODIhedral] The RTF reader AUTOgenerate command syntax: AUTOgenerate [ ANGLes ] [ DIHEdrals ] [ PATChes ] [ NONE ] [NOANgles] [NODIhedrals] [NOPAtches] Note that autogenerate is not the default and must be requested. The specifications in the PATCH command override the specification in the RTF. Files Modified: source/charmm/iniall.src source/fcm/rtf.fcm source/gener/genpsf.src source/gener/modpsf.src source/io/rtfio.src .............................................................................. 4.2.2 New MANTIM STATe and HEAViside Functions CORRECL has two new operations, STATe and HEAViside. MANTIME name { STATe minv maxv } ! Q(t) = 1 (minv < Q(t) < maxv), else = 0 { HEAViside } ! Q(t) = 0 (when Q(t)<0), else = 1 Files Modified: source/correl/correl.src source/correl/mantim.src .............................................................................. 4.2.3 New Crystal type "RECTangular" A new crystal type "RECTangle" has been added to the CRYStal DEFIne command. This new type is the same as "ORTHorhombic", except that the ratios of the unit cell dimensions are fixed. The total volume may change. RECTangular - a/A = b/B = c/C and alpha = beta = gamma = 90.0 degrees. (Note: A,B,C are the initial a,b,c values) (example: 50.0 40.0 30.0 90.0 90.0 90.0 ) (volume = c*b*a) (degrees of freedom = 1) The undocumented ISOTropic option in the DYNAmics command has been removed, since it never worked properly and the new code make its intended function obsolete. Files Modified: source/dynamc/dynamvv2.src source/dynamc/prssre.src source/energy/energy.src source/fcm/image.fcm source/image/crystal.src source/image/eimage.src source/image/xtlfrq.src source/mc/mcener.src source/mc/mcmini.src source/minmiz/abner.src source/minmiz/conjug.src source/minmiz/egrad1.src source/minmiz/nraph.src source/minmiz/powell.src source/minmiz/steepd.src source/minmiz/tndriv.src source/misc/testch.src source/pert/epert.src .............................................................................. 4.2.4 Add SCALAR WRITE Command There is a new SCALar WRITe command to output data arrays in CHARMM. File Modified: source/manip/scalar.src .............................................................................. 4.2.5 New Expanded Fast Routines and Fast Control A new set of FAST routines have been developed. These have been written as a single routine that becomes expanded into many "cleaned" routines that are free of unnecessary code (e.g. IF tests). Use of these routines will significantly improve performance. FAST OFF ! Use the slow full-feature routines FAST GENEric ! Use the old "fast" routines (that are no longer fast) FAST ON ! Use best routine available for current calculation FAST EXPAnd ! Use new expanded routines Files Added: source/nbonds/enbaexp.src source/nbonds/enbgexp.src Files Modified: source/charmm/miscom.src source/energy/enefscal.src source/energy/energy.src source/energy/intere.src source/fcm/fast.fcm source/image/eimage.src source/mbond/mbdiag.src source/mc/mcimge.src source/misc/genetic.src source/nbonds/enbond.src source/pert/epert.src source/pert/icfcnf.src source/pert/icpert.src source/pert/tsme.src source/vibran/vibblock.src .............................................................................. 4.2.6 Inline ERFC Code for Faster Ewald The ERFC lookup table code has been simplified (fewer operations) and now works with a zero distance (good for DRUDE), which eliminated a pair of inner loop IF tests. Files Modified: source/cff/ewald_cff.src source/fcm/ewald.fcm source/mmff/enbscalar_mm.src source/nbonds/ewald.src source/nbonds/nbutil.src .............................................................................. 4.2.7 Improve the PRESsure INSTantaneous Command. The PRESsure INSTantaneous command now works with a zero temperature. Additional virial and pressure data is printed. Files Modified: source/dynamc/prssre.src .............................................................................. 4.2.8 NUMLP Set ?NUMLP variable for the number of lonepairs. Files Modified: source/dynamc/lonepair.src source/gener/psfsum.src .............................................................................. 4.2.9 PRNLEV Control Change PRNLEV values to improve print control. Files Modified: source/energy/printe.src source/image/images.src source/nbonds/heurist.src source/nbonds/nbonds.src .............................................................................. 4.2.10 Energy Call Control To facilitate routine selection in ENBOND, a universal flag, QGCMC, has been added. (Aaron, would you check on this? mctest02.inp seems to generate energy blow-out. I fixed part of it, yet does not work correct.) Files Modified: source/charmm/iniall.src source/fcm/gcmc.fcm .............................................................................. 4.2.11 Move ACE Code The ACE common block code was added to a structured data file in fcm/inbnd.fcm (which violates the rules governing structured data files). It also inhibited further development of inbnd.fcm. The ACE arrays have been moved to the file fcm/ace.fcm. File Added: source/fcm/ace.fcm Files Modified: source/charmm/iniall.src source/dynamc/dcntrl.src source/fcm/inbnd.fcm source/io/mainio.src source/manip/cstran.src source/manip/scalar.src source/mc/mc.src source/mc/mcace.src source/mc/mcener.src source/nbonds/ace.src source/nbonds/nbonds.src source/nbonds/nbutil.src source/util/selcta.src .............................................................................. 4.2.12 Set KBT and expand format statement The variable KBT was not properly set. File Modified: source/dynamc/dcntrl.src .............................................................................. 4.2.13 Optional Exit for SHAKE Error Added an optional exit for SHAKE tolerance exceeded errors. By setting BOMLEV to 1 or more, you can make CHARMM die on this error. File Modified: source/manip/shake.src .............................................................................. 4.2.14 Fix bugs/problems with ##IF usage There were a number of ##IF control problems. The use of the inline precompile selection: !##IF CHEQ is wrong. The correct syntax is !##CHEQ Also, spaces within ## line identifiers is not allowed. Files Modified: source/fcm/param.fcm source/fcm/traj.fcm source/fcm/vibblock.fcm source/image/upimag.src source/vibran/vibran.src .............................................................................. 4.2.15 Fix QECONT and ATOMON Bugs The use of QECONT and filling the ECONT array was wrong. File Modified: source/misc/eef1.src .............................................................................. 4.2.16 Fix SCALAR READ Parallel Bug There was a bug in the SCALar READ command when used in a parallel calculation. This has been fixed. File Modified: source/manip/scalar.src .............................................................................. 4.2.17 Fix Compile and Timer Bugs in SPAS Code The SPAS code would not compile properly. It still does not work and it consumes massive quantities of memory... The code should be reworked or removed. Files Modified: source/energy/energy.src source/io/parmio.src .............................................................................. 4.2.18 NUMBERs and CONSTANTs The use of real*8 numbers in calling sequences has been corrected. This is non-standard and causes problems on some compilers. Also, I made an effort to make the use of constants more consistent. For example, I replaced instances of 332.0... with the parameter CCELEC in fcm/consta.fcm. Files Modified: source/dynamc/dynamcv.src source/dynamc/dynamln.src source/dynamc/dynamvv.src source/dynamc/dynamvv2.src source/energy/ecmap.src source/energy/gbmv.src source/energy/gbsw.src source/fcm/consta.fcm source/io/coorio.src source/io/rtfio.src source/mbond/mbback.src source/mc/mcio.src source/misc/genetic2.src source/nbonds/pmeutil.src source/util/random.src .............................................................................. 4.2.19 Fix Spelling File Modified: source/fcm/rndnum.fcm .............................................................................. 4.2.20 QM/MM Interface Using the Q-Chem Quantum Chemistry Package Added a QM/MM interface to Q-Chem using the GAMESS-US and GAMESS-UK interface as a model. Q-Chem and CHARMM communicate via files rather than common blocks which means the two programs live separately. The Q-Chem QM/MM interface was also extended to function with the highly parallel Replica/Path and the Nudged Elastic Band Methods. Files Modified: install.com source/charmm/charmm_main.src source/charmm/iniall.src source/energy/energy.src source/fcm/gamess.fcm source/fcm/replica.fcm source/fcm/path.fcm source/gener/update.src source/gukint/gukini.src source/nbonds/nbonda.src source/nbonds/nbonds.src source/quantum/addlnat.src .............................................................................. 4.2.21 Q-Chem Input Generation Added the ability of zmat.src to write out Q-Chem input files. This occurs in the same way that the Gausssian interface is used. File Modified: source/misc/zmat.src .............................................................................. 4.2.22 Off-Path Optimization and Simulation The new EPATHO procedure was added which allows a user to run orthogonal simulations at each replica/bead/point along a replica or NEB path. These simulations are used to compute the free energy of reaction. Curvature corrections to these simulations were also added, but have not been throughly tested to determine their usefulness. Files Modified: source/energy/epath.src source/fcm/path.fcm source/fcm/replica.fcm source/fcm/gamess.fcm ------------------------------------------------------------------------------ 4.3 Developments at NIC-Slovenia Names : Milan Hodoscek Email Address : milan@cmm.ki.si Institution : National Institutes of Chemistry, Ljubljana, Slovenia Date : June 30, 2004 4.3.1 New version of GAMESS supported New version of GAMESS (19 May 2004 (R3)) is interfaced with CHARMM. New features include Fragment MO, NMR shifts, improved DFT convergence and many others. Files Modified: doc/gamess.doc source/fcm/gamess.fcm source/flucq/fluqqmmm.src source/gamint/blur.src source/gamint/ddi.src source/gukint/gukini.src source/gamint/gamess/{grd2a.src,int1.src,mccas.src,prplib.src, prppop.src,prpel.src,unport.src,zunix.c, gamess.src,grd1.src,guess.src,inputb.src, inputc.src,iolib.src,scflib.src,rhfuhf.src, inputa.src} .............................................................................. 4.3.2 Parallelization of DRUDE, LONEPAIR and VV2 integrator DRUDE and VV2 integrator are parallelized. For efficiency also LONEPAIR had to be parallelized. Files Modified: source/gener/drude.src source/manip/holonom.src source/machdep/paral1.src source/energy/ediff.src source/energy/energy.src source/dynamc/lonepair.src source/dynamc/dcntrl.src source/dynamc/dynamvv2.src source/dynamc/dynamcv.src source/dynamc/dynamc.src source/dynamc/dynio.src source/nbonds/nbonds.src .............................................................................. 4.3.3 Parallelization of SSBP and GSBP Spherical Solvent Boundary Potential method was parallelized. General Solvent Boundary Potential method parallelization started. Currently works in parallel environment, but not parallelized yet. Files Modified: doc/parallel.doc source/energy/energy.src source/misc/ssbp.src source/misc/gsbp.src source/misc/pbeq2.src .............................................................................. 4.3.4 Enhancements for GRAPE interface Several enhancement in the GRAPE interface. Files Modified: source/nbonds/grape.src .............................................................................. 4.3.5 Fix and extension of SOLANA The calculation of rotational correlation in SOLANA was wrong. It is fixed and extended to use Legendre polynomials of second order, by introducing new command keyword RLP . Files Modified: doc/corman.doc source/charmm/iniall.src source/correl/solana.src source/fcm/corman.fcm .............................................................................. 4.3.6 Parallelization of RANDOM code and fix for NEWRNG Parallelization of standard RANDOM number generator. Also the usage of CLCG generator was fixed at the same time. Files Modified: source/charmm/iniall.src source/correl/solana.src source/dynamc/dcntrl.src source/dynamc/dynamc.src source/energy/gbmv.src source/fcm/clcg.fcm source/fcm/parallel.fcm source/manip/intcor.src source/mc/mc.src source/misc/genetic.src source/misc/pbeq.src source/util/clcg.src source/util/random.src source/vibran/vibran.src .............................................................................. 4.3.7 Fixes for various ## keywords Some of the ## keywords would not allow compilation of CHARMM if excluded from pref.dat. New set of fixes for this are in: Files Modified: source/energy/eintern.src source/energy/energy.src source/nbonds/enbfast.src source/nbonds/nbonds.src source/mc/mcener.src ------------------------------------------------------------------------------ 4.4 Harvard Developments Name : Robert Petrella Email Address : petrella@tammy.harvard.edu Institution : Harvard University Date : June 29, 2004 4.4.1 BYCC Enhancements 1) SUBROUTINE NBNDCC made into two routines: NBNDCCF --for fastest execution, stripped of options NBNDCCO --for compatibility with various options. The latter routine is the one where developers should add functionality to BYCC. 2) BYCC made compatible with heuristic updates (for both atom-based and group-based calculations) 3) Changes in logic for calling cluster-generating routine (SUBROUTINE CLSBYRULE). A) Clusters are made by default prior to updating the list if "BYCC" is specified in the update command (unless they have previously been made). B) An informed guess is made as to whether the connectivity or the topology of the system has changed before calling NBNDCC. If a change is detected, the clusters are remade prior to non-bonded list generation. This is done because changes in the connectivity may necessitate the formation of new clusters, and significant changes in the topology almost certainly do. (changed subroutine: GTNBCT) 4) added global bycc QLBYCC in actclus.fcm to indicate BYCC has been specified in non-bonded command (for use in other parts of charmm). 5) minor modifications in eef1 code for compatibility with BYCC 6) BYCC made compatible with TSM 7) clustering algorithm made faster (now clusters roughly 1,000,000 atoms/s on 2.66 GHz Pentium 4) Files Modified: source/charmm/iniall.src source/nbonds/nbndcc.src source/nbonds/nbonds.src source/nbonds/nbutil.src source/misc/eef1.src source/fcm/actclus.fcm ------------------------------------------------------------------------------ 4.5 Developments at University of Vienna Developers : Tibor Rudas, Stefan Boresch and Othmar Steinhauser Email Address : tibi@mdy.univie.ac.at and stefan@mdy.univie.ac.at Institution : Molecular Dynamics and Biomolecular Simulation group at the Department of Theoretical Chemistry and Molecular Structural Biology, University of Vienna Date : June 23, 2004 4.5.1 COM Centering in COOR DIPO When calculating the dipole moment of a set of atoms carrying a total net charge the coordinates of this set are recentered to the center of geometry of these atoms. Adding a MASS keyword to the invocation (which is present for many other COOR commands) allows the center of mass as locus of recentering. Both this keyword and the recently introduced OXYZ keyword (which avoids this recentering unsing the absolute coordinates) are also introduced into the DIPO series of CORREL. All calls to CDIPOLE are updated but where no use/parsing of a MASS keyword is provided/necessary .FALSE. and 0 are passed instead of MASS and AMASS (solana.src) Files Modified: source/correl/correl.src source/correl/anacor.src source/correl/solana.src source/manip/corman.src Modified Documentation: doc/corman.doc doc/correl.doc .............................................................................. 4.5.2 SHELLs of Different Thickness Instead of the (still supported) keyword SHTH which defines a single thickness used for all shells generated shells of different thickness can be defined by using the new SHBO keyword. Following this keyword should be NSHL real numbers which represent the boundaries between the single shells (so these numbers should all be in increasing order). Files Modified: source/fcm/shell.fcm source/correl/shell.src source/charmm/iniall.src Modified Documentation: doc/shell.doc Modified Testcase: test/c31testshell.inp ------------------------------------------------------------------------------ 4.6 Karolinska Institute Fixes and Enhancements Developer : Lennart Nilsson Email Address : Lennart.Nilsson@biosci.ki.se Institution : Karolinska Institute Date : July 1, 2004 1) Added possibility to compute water-dipole rotational correlation fcn (P1 or P2), also works in shell defined by RSPIN & RSPOUT Files Modified: source/correl/solana.src doc/corman.doc test/c27test/solanal2.doc (seemed unnecessary to add another file for this) 2) Set CHARMM variable CFNORM to the normalization used in CORREL CORFUN, and added possibility to supply the normalization value to be used by CORFUN; useful to get correctly normalized partial correlation fcns. Files Modified: source/correl/solana.src source/correl/correl.src source/correl/corfun.src doc/correl.doc doc/subst.doc 3) Added substitution variables CPUTIME and ELATIME Files Modified: source/machdep/machutil.src doc/subst.doc 4) Corrected documentation of defaults for CORREL CORR (Legendre pol. P0 is a constant..) File Modified: doc/correl.doc ------------------------------------------------------------------------------ 4.7 Implicit Membrane Model 1 (IMM1) and EEF1 update Developer : Themis Lazaridis Email Address : tlazaridis@ccny.cuny.edu Institution : City College of New York/CUNY Date : May 17, 2004 IMM1 is an extension of the EEF1 implicit solvation model to lipid membranes. The membrane hydrocarbon interior is modeled as cyclohexane, with a smooth, monotonic transition between the interior and the aqueous region. Electrostatic screening is modeled using a modified distance-dependent dielectric function (eps=r**l, where l depends on the position of the two atoms with respect to the membrane. In addition, updated parameters for EEF1 are provided (referred to as EEF1.1) and also a version of EEF1 based on CHARMM22. Files Added: support/aspara/toph19_eef1.1.inp support/aspara/param19_eef1.1.inp support/aspara/top_all22_prot_eef1.inp support/aspara/top_all22_prot_eef1.1.inp support/aspara/solvpar22.inp Files Modified: source/fcm/eef1.fcm source/energy/energy.src source/energy/intere.src source/misc/eef1.src source/nbonds/enbfast.src support/aspara/solvpar.inp Modified Documentation: doc/eef1.doc ------------------------------------------------------------------------------ 4.8 Multiple Gaussian Job Input Generation Developers : Victor Anisimov and Alex D. MacKerell, Jr. Email Address : victor@outerbanks.umaryland.edu Institution : University of Maryland Date : June 28, 2004 Gaussian program uses "--link1--" string to separate multiple jobs. On the other hand hyphen symbol is treated by CHARMM as line continuation instruction. As a result, CHARMM was merging the "--link1--" string, which is a Gaussian instruction, with the next line. A correction has been made to instruct the CHARMM command line reader to turn off the line continuation mechanism for the moment of Gaussian file generation. The command line parser ignores hyphen symbol when the parser is invoked from G94INP routine. Logical variable USEHYP is introduced for this purpose. USEHYP set to .TRUE. enables line continuation mechanism (CHARMM default). USEHYP set to .FALSE. disables line continuation (call from G94INP routine). File Modified: source/util/parse.src ------------------------------------------------------------------------------ 4.9 SASA Changes and Enhancements Developers : Urs Haberthuer and Amedeo Caflisch Email Address : atlantis@bioc.unizh.ch, caflisch@bioc.unizh.ch Institution : University of Zurich, Switzerland Date : July 3, 2004 The keywords to modify the SASA parameters from within the CHARMM input file were changed for clearer usability. This concerns the keywords to modify the radii, probabilistic parameters, and connectivity parameters for the surface calculations. The keywords to modify the surface-tension like solvation parameters were also changed. See the SASA documentation for the new syntax. Two keywords, INFX and SURF, were added. INFX causes the fixed exclusion atoms to be considered for the surface calculations. (By default, they are neglected for historical reasons. See the SASA documentation for more detailed explanations.) SURF causes the approximated atomic solvent accessible surface areas to be stored in WMAIN. Lazaridis and coworkers introduced the CHARMM atom type CR in param19_eef1.inp and toph19_eef1.inp. This atom type is now considered explicitly in the SASA code although all the SASA parameters for CR are identical to those for the CHARMM atom type C. Some code cleanup and adjustments of the documentation within the SASA code were performed. Files Modified: doc/sasa.doc source/fcm/sasa.fcm source/misc/sasini.src source/misc/sasene.src ============================================================================== [5] c31b2 Change Log The following are introduced and modified during the development period of August 15, 2004 through February 15, 2005. The developments are based on CHARMM c31b1. 5.1 TSRI Developments Developers : Mike Crowley and Charles L. Brooks, III Email Address : crowley@sripps.edu, brooks@scripps.edu Institution : The Scripps Research Institute Date : December 30, 2004 5.1.1 AFM Reset The reset command is added to afm. This permits the afm commend to be run in a loop with all data structures cleared in between. Files modified: doc/afm.doc source/energy/afm.src source/fcm/afm.fcm .............................................................................. 5.1.2 Clean-Up Some code cleanup and prefx labeling to make debugging easier. Files modified: source/energy/energy.src source/energy/eutil.src source/nbonds/enbfast.src source/util/prpref.src .............................................................................. 5.1.3 GBMV Bugfix We fixed a number of issues related to the performance and correctness of the GBMV code on serial and parallel machines. Files modified: source/energy/gbmv.src source/fcm/gbmv.fcm .............................................................................. 5.1.4 Huge Version HUGE keysord added to code and install.com for million atom systems. Files modified: install.com source/fcm/dimens.fcm source/fcm/heap.fcm source/fcm/stack.fcm .............................................................................. 5.1.5 Makefile Update Updated osx and gnu Makefiles Files modified: build/UNX/Makefile_osx build/UNX/Makefile_gnu .............................................................................. 5.1.6 Testcase Fixes Small fixes to test cases to correct poor design of some test cases and to add some functionality to others. Files modified: test/c31test/nebdiala.inp test/c31test/afm.inp test/c31test/nebdialapmf.inp test/c31test/off_path_simulation.inp .............................................................................. 5.1.7 Timing Fixes Timer debugging added to prevent double timing and other timing errors. Files modified: source/fcm/new_time.fcm source/util/new_timer.src .............................................................................. 5.1.8 NEWRNG Code Removed NEWRNG produced bad results, code removed to recover old results. Files modified: source/charmm/iniall.src source/correl/solana.src source/dynamc/dcntrl.src source/dynamc/dynamc.src source/dynamc/dynio.src source/energy/gbmv.src source/fcm/exfunc.fcm source/manip/intcor.src source/mc/mc.src source/misc/genetic.src source/misc/pbeq.src source/util/random.src source/vibran/vibran.src ------------------------------------------------------------------------------ 5.2 Developments at the National Institute of Health Developers : H. Lee Woodcock, Xiongwu Wu and Bernard R. Brooks Email Address : hlwood@nih.gov, wuxw@nhlbi.nih.gov and brb@nih.gov Institution : Laboratory of Biophysical Chemistry, NHLBI, NIH Date : December 30, 2004 5.2.1 SGLD Bugfix The local average velocities are accumulated twice in the SGLDW subroutine. File Modified: source/dynamc/sglds.src ------------------------------------------------------------------------------ 5.3 SASA Enhancements and Bugfixes Developers : Urs Haberthuer, Riccardo Pellarin, and Amedeo Caflisch Email Address : atlantis@bioc.unizh.ch, pellarin@bioc.unizh.ch, caflisch@bioc.unizh.ch Institution : University of Zurich, Switzerland Date : December 30, 2004 (1) New Surface Parameters A new keyword, NEWP, is added. It triggers the use of a new set of values for the radii, probabilistic parameters, and connectivity parameters. (These parameters are used to calculate atomic solvent accessible surface areas within the SASA module.) The new set was derived in 2002 by Haberthuer and it gives a better correlation with exact analytical surfaces than the original Hasel and Still surface parameters. (2) Images Image lists are now supported in SASA. This allows to use SASA for simulations with periodic boundary conditions. (3) Documentation The SASA documentation itself and also the documentation within the SASA code is updated. (4) Miscellaneous The atom type C1ES is now supported by SASA and the format of the output of the SASA module slightly changed. (5) Bugfixes The SASA module did not work correctly for simulations with constant pressure since the call to the subroutine SASENE in the subroutine ENERGY was at the wrong place. This has been corrected. An empty nonbond exclusion pair list could crash the SASA module. This bug is now fixed. Files Modified: doc/sasa.doc source/energy/energy.src source/fcm/sasa.fcm source/misc/sasini.src source/misc/sasene.src ------------------------------------------------------------------------------ 5.4 Karolinska Institute Fixes and Enhancements Developer : Lennart Nilsson Email Address : Lennart.Nilsson@biosci.ki.se Institution : Karolinska Institute Date : December 30, 2004 (1) SINGLE precision version cannot be created due to use of non-generic forms of intrinsic functions in several routines. I updated prefx.src with these, and also had to make some changes to source files. CHARMM created with SINGLE in pref.dat is however NOT fully functional. Timing routines (elapsed time) do not work, parallel jobs hang (I only tried socket communication), charges do not sum up correctly at psf generation; there may be other problems as well. It would be useful to have this working correctly again. Files modified: tool/prefx.src source/adumb/eadumb.src source/emap/emapdock.src source/util/selcta.src (2) Set values of BOMLEV, IOLEV, PRNLEV variables at startup, corrected a typo in FTCPWRT call Files modified: source/charmm/charmm_main.src (3) Allow echeck to be turned off by specifying a negative value; not documented. Files modified: source/dynamc/dynamc.src (4) Some format changes to allow for large step numbers to print correctly. Files modified: source/dynamc/dynsub.src (5) Format change to allow larger heap allocations to be correctly reported. Files modified: source/machdep/space.src