Dear Netters; There is now a bulletin board service (bbs) for users of the CHARMM (Chemistry at HARvard Macromolecular Mechanics) modeling software. This bbs will be dedicated to the applications of charmm and the science behind it. --> This service will be available to anyone who uses charmm. <-- --> It is designed so that users can share charmm information, <-- --> or scripts, with others quickly and efficiently via e-mail. <-- ----- --> So far, several leading industrial charmm user groups have aggreed <-- --> to participate. These include; NIH, IBM, ELI-LILLY, Merck, Exxon,<-- --> Polygen/MSI etc. Several academic institutions as well. <-- +++++++++++++++++++++++++++++++++++++++++++++++++++++++++ +The charmm bbs will be called: + + + + charmm-bbs@emperor.harvard.edu + + + +Any personal msgs and REQUESTS TO JOIN can be sent to: + + + + charmm-bbs-request@emperor.harvard.edu + +++++++++++++++++++++++++++++++++++++++++++++++++++++++++ ----------------- Any charmm related questions/comments or scripts you would like to share, are simply posted to the bbs in the form of e-mail. All subscribers will recieve all e-mail that is sent to the bbs. For example: mail charmm-bbs@emperor.harvard.edu subject/ body of letter etc.. Stay tuned as a major release of CHARMm (CHARMm 22) from Polygen/MSI is eminant. Regards, Donovan Chin ================================================== Angelo Rossi rossi@watson.ibm.com Hello: With respect to the charmm-bbs I have the following comments/suggestions: 1) When it starts off, I think their should be some goodie about CHARMm to offer, i.e. hints, sample scripts ... something along these lines. 2) From the beginning the charmm-bbs must provide equal balance to both protein and polymer simulations. This will clearly differentiate it from other bbsesses that I have seen. All others seem to emphasize only protein modelling. Of course, one could argue that most principles and appends will apply equally to both areas, but there is a tendency to be turned off, if the information is not within an area of interest. Besides, CHARMm has fantastic potential for both polymer and protein modelling as we all know. 3) I would also try to raise the level of the charmm-bbs by making sure there is very high scientific content in the questions, appends and discussion (This will be the most difficult challenge to the charmm-bbs, in my opinion) Questions which are in the noise level (I realize this is subjective.), as I would say, should be avoided. These are questions that are functional or mechanical. I personally would like to see a dialogue or diatribe on constant pressure MD and its consequences, i.e. how it differs from the NVE ensemble and what are the physical implications. This is a typical area that is new and different in CHARMm. In my opinion this exemplifies the type append for the charmm-bbs. The ideal goal would be to have the charmm-bbs at the highest level of science at all times (I am becoming repetitive.) 4) I feel that Quantum Chemistry should not be included in a direct way but but could creep into the charmm-bbs indirectly, e.g. in terms of how it is used to derive force field parameters, etc. Best regards, A. R. Rossi ibm ---- Received: from BROWNVM.BROWN.EDU by brownvm.brown.edu (IBM VM SMTP R1.2.1MX) with BSMTP id 7817; Wed, 10 Jun 92 09:59:21 EDT Received: from BROWNVM (BUSATH) by BROWNVM.BROWN.EDU (Mailer R2.08 R208004) with BSMTP id 3475; Wed, 10 Jun 92 09:59:20 EDT Date: Wed, 10 Jun 92 09:56:44 EDT From: David Busath Subject: Re: large forces To: Donovan Chin In-Reply-To: Your message of Tue, 9 Jun 92 16:16:49 -0400 Status: RO I have a student, Jim Fukuda, who has worked out how to display the forces during a dynamics run using the fourth parameter with quanta and dynamics animation. He is in Hong Kong and I haven't been able to reach him the last few days, but if he becomes communicative by email from Hong Kong, I will forward your note to him. Meanwhile, you could probably work it out - its not too hard, but does require some file manipulating. --- From polygen!phoenix!mrp Wed Jun 10 12:02:08 1992 Received: from polygen.UUCP by BU.EDU (1.99) Wed, 10 Jun 92 12:03:47 -0400 Received: from phoenix.polygen by polygen.com (4.0/SMI-4.0) id AA02364; Wed, 10 Jun 92 09:11:47 EDT Received: by phoenix.polygen (4.1/SMI-4.1) id AA04507; Wed, 10 Jun 92 09:11:54 EDT Date: Wed, 10 Jun 92 09:11:54 EDT From: polygen!phoenix!mrp (Mike Pear) Message-Id: <9206101311.AA04507@phoenix.polygen> To: dchin@emperor.harvard.edu Subject: Re: large forces Status: R Donovan, Look in the charmm22 documentation in "corman.doc". Line 320 gives an example for printing out large forces. By putting this in a loop with a "TRAJ READ" command, you can get large forces at each point during a trajectory. You can limit this set of forces to those acting on the zinc by using the "INTEraction" command (see energy.doc) where one of the atom selections specifies the zinc. --- From WSONNEN@EAGLE.WESLEYAN.EDU Thu Jun 11 10:32:29 1992 Received: from EAGLE.WESLEYAN.EDU by EAGLE.WESLEYAN.EDU (PMDF #12650) id <01GL32OFC8KW90N5BR@EAGLE.WESLEYAN.EDU>; Thu, 11 Jun 1992 10:32 EDT Date: 11-JUN-1992 10:25:29.09 From: WSONNEN@EAGLE.WESLEYAN.EDU Subject: CHARMm IMPATch (HELP!) To: charmm-bbs@emperor.harvard.edu Message-Id: <01GL32OFC8KW90N5BR@EAGLE.WESLEYAN.EDU> Status: RO (1) Is there anyone on this network who has used the IMPAtch command in CHARMm? At present I have no documentation on how to use this command. Any help whold be appreciated. I am not a subscriber to this network (yet), please respond to me directly. Thanks you, Wayne Sonnen wsonnen@eagle.wesleyan.edu P.S. Can someone tell me how I can subscribe to this network. I am a (beginning) CHARMm user. Here's a note to test out the system: We recently found out that in order to use the FREE command on a trajectory file from a previous run, it is necessary to assure that the non-bonded list is generated in FREE just as it was during the dynamics computation. This requires that one read in the initial configuration file and execute UPDATE before the FREE and that one include the keyword INBFRQ with the FREE command to identify the nonbonded list update frequency used to generate the dynamics trajectory. Otherwise, the nonbonded list will not be the same as that used during the dynamics computation and FREE will give different energies than it would have if executed immediately after the DYN command. -- From dchin@emperor.harvard.edu Tue Jun 9 16:16:37 1992 Date: Tue, 9 Jun 92 16:16:49 -0400 From: dchin@emperor.harvard.edu (Donovan Chin) To: charmm-bbs@emperor.harvard.edu Subject: large forces Status: OR Does anyone know how you can identify large forces that are occuring in a dynamics simulation? I have a zinc ion that is jumping out of its binding pocket during the first .1 ps of a langevin simualtion and i would like to identify the offending force of interaction. Donovan Chin --- From dchin@emperor.harvard.edu Wed Jun 10 12:34:18 1992 Date: Wed, 10 Jun 92 12:33:56 -0400 From: dchin@emperor.harvard.edu (Donovan Chin) To: charmm-bbs@emperor.harvard.edu Subject: finding large forces. Status: OR Thanks to Mike Pear and old (real old) Bernie Brooks notes, the following script helps in finding large forces in a dynamics file; *FORCES.STR * open read unit 32 file name camgly_nowat_s15_a.dcd traj iread 32 nfile 10 nread 1 ener set 1 1 traj read label loop traj read inte sele resname zn end sele resname glu end !or whatever COOR FORCE COMP ! copy forces to the comparison coordinates COOR DIST WEIGH COMP ! put magnitudes in the weighting array. PRINT COOR COMP SELE PROP 1 .GT. 5.0 END ! print atoms with large incr 1 by 1 if 1 le 4 goto loop return --- CHARMm users: I understand that Bernard Brooks at the NIH offers a class on using CHARMm. I you have a copy of his class notes, I would be very interested in obtaining a copy. I am, of course, willing to compensate you for the cost of xeroxing and mailing, and would be forever indebted to you for your time. Please contact me directly to let me know if you are willing to do this for a fellow (beginning) CHARMm user. Thank you, wayne sonnen wsonnen@eagle.wesleyan.edu --- From brbrooks@helix.nih.gov Thu Jun 11 11:57:19 1992 Return-Path: Received: from polygen.com by fred.polygen.com (3.2/SMI-3.2) id AA04781; Thu, 11 Jun 92 11:57:17 EDT Received: from connection.prospect.com by polygen.com (4.0/SMI-4.0) id AA03337; Thu, 11 Jun 92 11:57:26 EDT Received: from thurifer.harvard.edu by connection.prospect.com (AIX 1.3/4.03) id AA20845; Thu, 11 Jun 92 11:56:43 -0400 Received: by helix.nih.gov (5.64/helix-1.0) id AA08625; Thu, 11 Jun 92 11:57:59 -0400 Date: Thu, 11 Jun 92 11:57:59 -0400 From: brbrooks@helix.nih.gov (Bernie Brooks) Message-Id: <9206111557.AA08625@helix.nih.gov> To: WSONNEN@EAGLE.WESLEYAN.EDU, charmm-bbs@emperor.harvard.edu Subject: Re: BROOKS'S NOTES Status: R I have been distributing my CHARMM course notes free of charge to anyone with CHARMM, or interested in CHARMM. The copying is done at a slowe rate, so don't expect an immediate response, but anyone interested in these notes should contact me. There is no way the I can accept any compensation for this service. I you are interested in attending some classes where this material is presented, contact me. Bernie Brooks, NIH --- From arne@mango.mef.ki.se Wed Jun 17 07:30:48 1992 Received: by mango.mef.ki.se (AIX 3.1/UCB 5.61/4.03) id AA20911; Wed, 17 Jun 92 13:30:08 +0100 Date: Wed, 17 Jun 92 13:30:08 +0100 From: arne@mango.mef.ki.se (Arne Elofsson) Message-Id: <9206171230.AA20911@mango.mef.ki.se> To: charmm-bbs@emperor.harvard.edu Subject: Benchmarks. (Someone have to use this bulletinboard) Cc: arne@mango.mef.ki.se Status: R I would be interested in benchmarks of different features of CHARMM, I suggest that a simple division would be in some categories. 1. Minimization of a protein (BPTI ?) 2. Dynamics of a protein in vacuum (BPTI ?) 3. A full periodic boundary simulation (constant pressure ?) 4. A smaller dynamic system (dipeptide) It would be interesting of benchmarks between difference machines (RS/6000 HP 750 SUN/10 CRAY YMP CONVEX C2/C3 CM2 CM5 Flamingo Crimson ...) It would also be interesting of comparison between different versions of CHARMM (we noted that CHARMM20 is noticeable faster than 22 for "plain" dynamics. A comparison of CHARMM with other (simpler and faster) dynamics programs as well as with AMBER GROMOS and other commercial programs. This should give important information to the developers of CHARMM. Arne Elofsson arne@mango.mef.ki.se --- Good morning - I am having a problem implementing a MD technique and I am looking for some help in how to proceed. Recently there have been several papers by Boryeu Mao at Upjohn regarding mass weighted molecular dynamics. This technique is used to perform a conformational search through the space of proper dihedral angles for peptides or protein fragments. It involves weighting the atomic masses of individual atoms by some arbitrary factor to blast through potential energy barriers in the course of the MD simulation. References for the technique are listed below: 1. Mao, B. & Friedman, A. R. Biophys. J. 58, 803-805 (1990). 2. Mao, B. Biophys. J. 60, 611-622 (1991). 3. Mao, B., Maggiora, G. M. & Chou, K. C. Biopolymers 31, 1077-1086 (1991). 4. Mao, B. Biophys. J. 60, 966-973 (1991). Now to my question: In implementing the technique, one needs to apply harmonic constraints on all internal coordinates except for proper dihedral angles. That's the problem for me. I know that the CONS IC command can be used for the bond lengths and bond angles but what of the improper dihedrals? Does anyone have a suggestion for applying constraints on just the improper dihedral angles? That is, other than resetting the force constant individually for each angle. I have the Polygen/Molecular Simulations version of CHARMM which was sent out with Quanta 3.3. Thanks in advance... Jeffrey Nauss -------------------------------------------------------------------------------- E-mail address: nauss@wrair-emh1.army.mil Mailing Address: Department of Gastroenterology Division of Medicine Walter Reed Army Institute of Research ATTN: MAJ Jeffrey L. Nauss, Ph.D. Washington, D.C. 20307-5100 Telephone: 202-576-3485 ---- Date: Thu, 6 Aug 92 11:04:39 EDT From: "Angelo R. Rossi" To: charmm-bbs%emperor.harvard.edu@relay.cs.net Subject: Question Regarding Time Series and Correlation Functions in CHARMm Status: RO Hello: I am using CHARMm to study various diffusant molecules in an alkane solvent such as pentane. Pentane has two dihedral angles capable of assuming various conformations. Before I study diffusion, I need to prepare the solvent by running dynamics to equilibrate the solvent so that the appropriate ratio of trans/gauche pentane is obtained. I want to look at that distribution by using the CHARMm Time Series and Correlation Function facility. The problem is that the "enter" command seems to allow the definition of only one dihedral angle at a time. I have 128x2 corresponding to a solvent of 128 pentane molecules each with two dihedral angles. The question then is there a way to use the CHARMm Time Series facility so that I obtain a single averaged (i.e. averaged over 128 pentane moleules) time series for each of two differen dihedral angles in pentane? The typical CHARMm script I am using follows. Thanks in advance for any suggestions you may have. Best regards, A. R. Rossi * CHARMM SCRIPT FOR ANNEALING OF LIQUID PENTANE * OUTU 6 BOMLEV -6 BANN ! READ RTF FILE FOR PENTANE read rtf card * pentane all atom * 20 1 MASS 3 HA 1.00800 ! Aliphatic or aromatic hydrogen MASS 10 CT 12.01100 ! Aliphatic carbon (tetrahedral) MASS 10 CT 12.01100 ! Aliphatic carbon (tetrahedral) AUTOGENERATE ANGLES DIHEDRALS DEFA FIRS NONE LAST NONE RESI pent 0.0 GROUP ATOM C1 CT -0.270 ATOM C2 CT -0.180 ATOM H3 HA 0.090 ATOM H4 HA 0.090 ATOM H5 HA 0.090 ATOM C6 CT -0.180 ATOM H7 HA 0.090 ATOM H8 HA 0.090 ATOM C9 CT -0.180 ATOM H10 HA 0.090 ATOM H11 HA 0.090 ATOM C12 CT -0.270 ATOM H13 HA 0.090 ATOM H14 HA 0.090 ATOM H15 HA 0.090 ATOM H16 HA 0.090 ATOM H17 HA 0.090 BOND C1 C2 BOND C2 C6 BOND C6 C9 BOND C9 C12 BOND C12 H15 BOND C12 H16 BOND C12 H17 BOND C9 H13 BOND C9 H14 BOND C6 H10 BOND C6 H11 BOND C2 H7 BOND C2 H8 BOND C1 H3 BOND C1 H4 BOND C1 H5 IC H3 C1 C2 C6 1.09 109.47 180.00 108.95 1.51 IC H4 C1 C2 C6 1.09 109.47 59.60 108.95 1.51 IC H5 C1 C2 C6 1.09 109.47 -59.60 108.95 1.51 IC H3 C1 C2 H7 1.09 109.47 60.32 109.47 1.09 IC H3 C1 C2 H8 1.09 109.47 -60.32 109.47 1.09 IC H3 C1 C2 H8 1.09 109.47 -60.32 109.47 1.09 IC C1 C2 C6 C9 1.51 108.95 180.00 108.95 1.52 IC H7 C2 C6 C9 1.09 109.47 -60.32 108.95 1.52 IC H8 C2 C6 C9 1.09 109.47 60.32 108.95 1.52 IC C1 C2 C6 H10 1.51 108.95 60.32 109.47 1.09 IC C1 C2 C6 H11 1.51 108.95 -60.32 109.47 1.09 IC C2 C6 C9 C12 1.51 108.95 180.00 108.95 1.52 IC H10 C6 C9 C12 1.09 109.47 -60.32 108.95 1.52 IC H11 C6 C9 C12 1.09 109.47 60.32 108.95 1.52 IC C2 C6 C9 H13 1.51 108.95 60.32 109.47 1.09 IC C2 C6 C9 H14 1.51 108.95 -60.32 109.47 1.09 IC C6 C9 C12 H15 1.52 108.95 -180.0 109.47 1.09 IC H13 C9 C12 H15 1.09 109.47 -60.32 109.47 1.09 IC H13 C9 C12 H16 1.09 109.47 +60.32 109.47 1.09 IC H13 C9 C12 H17 1.09 109.47 180.00 109.47 1.09 IC H14 C9 C12 H15 1.09 109.47 60.32 109.47 1.09 IC H14 C9 C12 H16 1.09 109.47 180.00 109.47 1.09 IC H14 C9 C12 H17 1.09 109.47 -60.00 109.47 1.09 IC C6 C9 C12 H16 1.52 108.95 59.60 109.47 1.09 IC C6 C9 C12 H17 1.52 108.95 -59.60 109.47 1.09 END READ PARA CARD * parameters for pentane * BOND CT CT 235.5 1.530 HA CT 330.0 1.100 THETA CT CT CT 60.0 113.0 HA CT HA 37.79 108.5 HA CT CT 44.91 109.5 PHI X CT CT X 0.157 3 0.0 ! NONB NBXMOD 5 ATOM CDIE SWITCH VATOM VDISTANCE VSWITCH - CUTNB 99.0 CTOFNB 98.0 CTONNB 97.0 EPS 1.0 E14FAC 1.0 WMIN 1.5 HA 0.100 -0.0045 1.468 CT 1.65 -0.0903 1.800 ! END DELE ATOM SELECT ALL END ! Now read monomer sequence for this structure. READ SEQUENCE CARD * PENTANE SOLVENT * 128 PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT PENT GENERATE SOLV SETUP ! READ STARTING COORDINATES OPEN READ CARD UNIT 7 NAME "pent6e3.crd" READ COORD IGNO UNIT 7 COOR STAT COOR TRANS XDIR ?XAVE YDIR ?YAVE ZDIR ?ZAVE FACT -1.0 - SELE ALL END COOR STAT SCALAR WMAIN = CHARGE SET 2 14.507675 SET 3 14.507675 SET 4 14.507675 SET 5 14.507675 DECREMENT 4 BY 2.0000 DECREMENT 5 BY 5.0000 SET 6 29.01535 READ IMAGES CARD * IMAGE TRANSFORMATION FILE * SCALE @6 @6 @6 IMAGE UP TRANSLATE 0.000 0.000 1.000 IMAGE DN DEFINE INVERSE UP IMAGE T3 TRANSLATE 1.000 0.000 1.000 IMAGE T4 DEFINE INVERSE T3 IMAGE T5 TRANSLATE 0.000 1.000 1.000 IMAGE T6 DEFINE INVERSE T5 IMAGE T7 TRANSLATE 1.000 1.000 0.000 IMAGE T8 DEFINE INVERSE T7 IMAGE T9 TRANSLATE 1.000 0.000 0.000 IMAGE T10 DEFINE INVERSE T9 IMAGE T11 TRANSLATE 0.000 1.000 0.000 IMAGE T12 DEFINE INVERSE T11 IMAGE T13 TRANSLATE 0.000 1.000 -1.000 IMAGE T14 DEFINE INVERSE T13 IMAGE T15 TRANSLATE 1.000 -1.000 0.000 IMAGE T16 DEFINE INVERSE T15 IMAGE T17 TRANSLATE 1.000 0.000 -1.000 IMAGE T18 DEFINE INVERSE T17 IMAGE T19 TRANSLATE 1.000 1.000 1.000 IMAGE T20 DEFINE INVERSE T19 IMAGE T21 TRANSLATE 1.000 -1.000 1.000 IMAGE T22 DEFINE INVERSE T21 IMAGE T23 TRANSLATE 1.000 1.000 -1.000 IMAGE T24 DEFINE INVERSE T23 IMAGE T25 TRANSLATE -1.000 1.000 1.000 IMAGE T26 DEFINE INVERSE T25 END PRINT IMAGES TRANSFORM ! UPDATE INBFRQ 25 CUTNB @3 CTONNB @4 CTOFNB @5 - ! VSHIFT SHIFT CDIE EPS 1.000000 IHBFRQ 0 IMGFRQ 25 CUTIM @2 ! ENERGY CALCULATION FAST 1 SHAKE OFF SKIPE NONE ENERGY - INBFRQ 25 CUTNB @3 CTOFNB @4 CTONNB @5 - VSHIFT SHIFT CDIE EPS 1.000000 - IHBFRQ 0 - IMGFRQ 25 CUTIM @2 FAST 0 COOR STAT IMAGE BYRES XCEN ?XAVE YCEN ?YAVE ZCEN ?ZAVE SELE SEGI SOLV END open read unit 51 file name pent6e4.dcd open write unit 20 card name aaaa.cor correl maxseries 1 maxtimesteps 500 maxatoms 5 enter aaaa torsion solv 2 c1 solv 2 c2 solv 2 c6 solv 2 c9 geometry trajectory firstu 51 nunit 1 mantime aaaa daver write aaaa unit 20 dumb time * title * STOP -- Date: 18-AUG-1992 22:36:28.46 From: WSONNEN@EAGLE.WESLEYAN.EDU Subject: CHARMM CRYSTAL BUILDER To: charmm-bbs Message-Id: <01GNQRRAR6848Y5S5F@EAGLE.WESLEYAN.EDU> Status: RO I would like to build a crystal in CHARMM and write out the coordinates of all images explicitly. I am using CHARMm21.3. As a test i have tried the following script; in the test script, i attempt to write-out coordinates for three images (according to the CHARMM notes, the inverses will not be written). The result i obtain, however, is two coordinate sets, the first of which is the origional coordinates that i read in as my assymetric unit. I don't know what the second set of coordinates are (although this wouldn't be too difficult to figure out--just keep one image and its inverse and see which one matches). My questions are: (1) how can i get CHARMM to write out all the coordinates for the images that i generate? (2) Can i generate a given image without explicitly specifing its inverse? Any responses should probably be sent over the network for other CHARMM users. Thank you, Wayne Sonnen wsonnen@eagle.wesleyan.edu The following is the script (after psf generation and coords are read): CRYSTAL DEFINE ORTHORHOMBI 24.8700 40.3800 66.2000 90.0000 90.0000 90.0000 CRYSTAL ZERO 24.8700 40.3800 66.2000 90.0000 90.0000 90.0000 CRYSTAL STRESS 0.000 0.000 0.000 0.000 0.000 0.000 CRYSTAL BUILD NOPER 3 CUTOFF 99.0 (1/2-X,-Y,1/2+Z) (-X,1/2+Y,1/2-Z) (1/2+X,1/2-Y,-Z) READ CRYSTAL CARD * DEFINITION OF CRYSTAL FORxxx.msf * SYMMETRY (X,Y,Z) (1/2-X,-Y,1/2+Z) (-X,1/2+Y,1/2-Z) (1/2+X,1/2-Y,-Z) END IMAGES 2 0 0 0 3 0 0 0 4 0 0 0 2 -1 0 0 3 0 -1 0 4 0 0 -1 END UPDATE INBFRQ 50 IMGFRQ 50 IHBFRQ 50 CUTIM 99.0 PRINT COOR IMAGE OPEN WRITE UNIT 88 CARD NAME TEST_CRYSTAL.CRD WRITE COOR IMAGE CARD UNIT 88 * STOP -- From dchin%emperor@harvard.harvard.edu Tue Aug 18 15:23:10 1992 Date: Tue, 18 Aug 92 15:16:27 -0400 From: dchin%emperor@harvard.harvard.edu (Donovan Chin) To: charmm-bbs%emperor@harvard.harvard.edu Subject: Charmm and quantum mechnics Status: OR Does anyone on this list have a charmm script using the Quantum mechanical/moleclular mechanics (QM/MM) module of charmm(22)? I would like to learn about this and the charmm syntax involved. I have the latest charmm lecture notes, unfortunatly it dose not contain any charmm syntax or script info. Thanx in advance --Donovan --- Date: Mon, 31 Aug 92 15:07:46 EDT From: polygen!momma!pqian (Peter Qian) Message-Id: <9208311907.AA08222@momma.polygen> To: charmm-bbs@bu.UUCP Subject: Ewald summation Status: RO Hi: Is there any one who is using Ewald summation in CHARMm? I have some problems in using them in CHARMm. I have made a box of 1,3-butanediol diacetate (40 molecules). Then I use the EWALD and KAPPA commands in CHARMm to calculate energy. However, the IMELec energy keeps change with the change of KAPPA value. That is, the energy does not converge. Any help is appreciated. Peter From varma@sfu.ca Wed Sep 9 16:24:26 1992 Received: from whistler.sfu.ca by emperor.harvard.edu via SMTP (920110.SGI/911001.SGI) for dchin id AA10288; Wed, 9 Sep 92 16:24:26 -0400 Received: from beaufort.sfu.ca by whistler.sfu.ca (5.65/SFU-2.0) id AA16898; Wed, 9 Sep 92 13:24:04 -0700 Received: by beaufort.sfu.ca (920330.SGI/911001.SGI) for charmm-bbs@emperor.harvard.edu id AA22427; Wed, 9 Sep 92 13:24:00 -0700 Date: Wed, 9 Sep 92 13:24:00 -0700 From: varma@sfu.ca Message-Id: <9209092024.AA22427@beaufort.sfu.ca> To: charmm-bbs Subject: Quanta Status: RO Does anyone out there have programs to read DCD and CSR files generated by Quanta ? Vikram Varma Simon Fraser University --- From varma@sfu.ca Wed Sep 23 16:41:22 1992 Date: Wed, 23 Sep 92 12:16:17 -0700 From: varma@sfu.ca To: charmm-bbs%emperor@harvard.harvard.edu Subject: Charmm21r3 Status: OR I have run into a problem while running a torsion search using charmm. The version of charmm that I am running is charmm21r3exe supplied by MSI. The command WRITE TITLE UNIT 20 * @1, @2, ?ENER does not write to the unit 20 but instead writes this line RDTITL> * 0.0, 220, 2.82199 to the output file as given in the command line ie charmm21r3exe < input.file > output.file Since the printout is large, I want to send it to /dev/null but keep the dihedral angles and energy. Does anyone know what the problem is? --- From arne@mango.mef.ki.se Thu Sep 24 06:47:23 1992 Date: Thu, 24 Sep 92 10:53:32 +0100 From: arne@mango.mef.ki.se (Arne Elofsson) To: charmm-bbs%emperor@harvard.harvard.edu Subject: Re: Charmm21a3 Status: OR At least in CHARMM20 and 22 (academic versions) is it necessary to finish the last line of the title with a single star (*) Write title unit 20 * @1, @2, ?ener * This should work I think yours arne --- From CSC001@UKCC.uky.edu Fri Sep 25 19:20:22 1992 Date: Fri, 25 Sep 92 15:47:52 EDT From: david ward To: charmm-bbs%emperor@harvard.harvard.edu Status: OR Is there a suggested value for epsilon to use with RDIE (in peptide simulations)? --- From dchin%emperor@harvard.harvard.edu Sat Sep 26 14:42:04 1992 Date: Sat, 26 Sep 92 14:42:14 -0400 From: dchin%emperor@harvard.harvard.edu (Donovan Chin) To: charmm-bbs%emperor@harvard.harvard.edu Subject: RDIE and epsilon. Status: OR I dont know anybody who uses a value other than 1 when using RDIE. Maby someone has used a value gt 1? Partial charges in charmm are parametized to a constant dielectric ? Epsilon (dielectric constant) is set to "r" when using RDIE. ie (qi*qj)/(4*pi*e0*r*r) where e0 is the permittivity of vacuum. references of interest: Loncharich, R. J., Brooks, B. R. "The Effects of Truncating Long-Range Forces on Protein Dynamics", Proteins, 6: 32-45, 1989. Mehler, E. L., Solmajer, T., "Electrostatic Screening in Molecular Dynamics Simulations", Protein Engineering, vol 4, 8: 911-917, 1991. ---- From varma@sfu.ca Tue Sep 29 22:43:58 1992 Received: from harvard.harvard.edu by emperor.harvard.edu via SMTP (920110.SGI/911001.SGI) for dchin id AA00985; Tue, 29 Sep 92 22:43:58 -0400 Received: by harvard.harvard.edu (5.54/a0.25) (for charmm-bbs@emperor) id AA18887; Tue, 29 Sep 92 22:43:31 EDT Received: from garibaldi.sfu.ca by whistler.sfu.ca (5.65/SFU-2.0) id AA27041; Tue, 29 Sep 92 19:43:26 -0700 Received: by garibaldi.sfu.ca (920330.SGI/911001.SGI) for charmm-bbs%emperor@harvard.harvard.edu id AA27374; Tue, 29 Sep 92 19:42:10 -0700 Date: Tue, 29 Sep 92 19:42:10 -0700 From: varma@sfu.ca Message-Id: <9209300242.AA27374@garibaldi.sfu.ca> To: charmm-bbs%emperor@harvard Subject: H-bonding Status: RO I have noticed while doing torsion searches that some conformations are stabilized by H-bonding, producing artificially low minima due to H-bonds that might not exist in solution. Is there anyway to shut off the H-bonding energy, while keeping the electrostatic term? Vik --- Date: Fri, 18 Sep 92 11:08:59 +0100 From: arne@mango.mef.ki.se (Arne Elofsson) Message-Id: <9209181008.AA04117@mango.mef.ki.se> To: charmm-bbs Subject: Stream handling Status: RO I think there is a BUG ! When you do a stream command to a file that not exists. I get the same ouput (Level 1 warning) until I have filled up my diskspace. Here is a (small) part of the log: CHARMM> stream ../stream/:rnt1defs.str VOPEN> Attempting to open::../stream/:rnt1defs.str:: OPNLGU> Unit 99 cannot be opened as ../STREAM/:RNT1DEFS.STR ***** LEVEL 0 WARNING FROM ***** ***** "OPEN" not possible. ****************************************** BOMLEV ( -1) IS NOT REACHED. WRNLEV IS 5 INPUT STREAM SWITCHING TO UNIT 99 RDTITL> STREAM ../STREAM/:RNT1DEFS.STR ***** LEVEL 1 WARNING FROM ***** ***** Title expected. ****************************************** BOMLEV ( -1) IS NOT REACHED. WRNLEV IS 5 RDTITL> No title read. CHARMM> STREAM ../STREAM/:RNT1DEFS.STR VOPEN> Attempting to open::../stream/:rnt1defs.str:: OPNLGU> Unit 99 cannot be opened as ../STREAM/:RNT1DEFS.STR ***** LEVEL 0 WARNING FROM ***** ***** "OPEN" not possible. This is produved from charmm22g1a Arne Elofsson --- Date: Wed, 07 Oct 92 12:47:57 EDT From: David Busath Subject: polarization To: CHARMM-BBS%emperor@harvard.harvard.edu Status: OR What is the status of polarization terms in the energy calculations? We have just received CHARMM22 and anticipated that it might allow a polarization energy to be included, but that doesn't appear to be the case. Are we missing something? From polygen!asteroid!ryszard Wed Oct 14 09:05:58 1992 Received: from BU.EDU by emperor.harvard.edu via SMTP (920110.SGI/911001.SGI) for dchin id AA14821; Wed, 14 Oct 92 09:05:58 -0400 Received: from polygen.UUCP by BU.EDU (1.99) Wed, 14 Oct 92 09:02:25 -0400 Received: from asteroid.polygen by polygen.com (4.0/SMI-4.0) id AA03953; Tue, 13 Oct 92 16:35:31 EDT Received: by asteroid.polygen (4.1/SMI-4.1) id AA11936; Tue, 13 Oct 92 15:41:03 CDT Date: Tue, 13 Oct 92 15:41:03 CDT From: polygen!asteroid!ryszard (Ryszard Czerminski X 285) Message-Id: <9210132041.AA11936@asteroid.polygen> To: dchin@bu.UUCP Subject: polarization Status: R Donovan, As far as I know there is no polarization term in current version of charmm (i.e. charmm22). Some time ago Bernie was mentioning polarization energy as one of the many items on his list 'to do in charmm', but I do not know what is the status of this. Ryszard -- From mmammen@gmwgroup Tue Oct 20 14:12:24 1992 Received: from gmwgroup.harvard.edu by emperor.harvard.edu via SMTP (920110.SGI/911001.SGI) for dchin id AA20930; Tue, 20 Oct 92 14:12:24 -0400 Message-Id: <9210201812.AA20930@emperor.harvard.edu> Date: 20 Oct 92 14:11:28 U From: "Mathai Mammen" Subject: Rotational Partition Function and Solvent To: charmm-bbs Status: RO Generally one ignores the effect of solvent on rotational configurational space. This seems intuitively resonable for small compact molecules, but does not seem so reasonable for larger more floppy molecules. I am specifically interested in the rotational (overall) entropy in solvent and am asking whether this differs significantly from gas phase rotational entropy calculated from a standard rotational partition function. If it does differ, how can one modify the partition function to account for solvent. Mathai From laaksone@csc.fi Tue Oct 20 16:38:05 1992 Received: from convex.csc.fi by emperor.harvard.edu via SMTP (920110.SGI/911001.SGI) for dchin id AA21173; Tue, 20 Oct 92 16:38:05 -0400 Received: by csc.fi id AA06953 (5.65c8+/IDA-1.4.4 for charmm-bbs@emperor.harvard.edu); Tue, 20 Oct 1992 22:36:59 +0200 Date: Tue, 20 Oct 1992 20:05:45 +0000 (GMT) From: Leif Laaksonen Subject: Bugs/problems and other questions To: charmm-bbs Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Status: RO Dear CHARMM fathers and mothers, I have been using the charmm22g1 version/release from Harvard soon for about a year. I'm mostly very happy with CHARMM and I still think that it is the best MM/MD program available but there are some questions which I have not received an answer to: (1) When CHARMM22 was released at the beginning of this year there was some malloc stuff in the IRIS version which had to be changed to be able to compile CHARMM22 with IRIX4. (2) I have not so far been able to use CHARMM22 with the periodic boundaries on our Cray-XMP. The program just bombs. (3) I also 'heard it through the grape wine' that there were problems with the periodic bounday conditions on the Convex as well. I have, however, had no problems on our Convex. I have reported all strange CHARMM22 behaviour to Harvard but I have never received a prompt reply about these bugs/problems. Does this mean that nobody else has seen these things and I'm regarded as a nuisance. Perhaps we are not supposed to report bugs at all. I just don't understand how we can develop CHARMM further if we can't report bugs and if we don't get some kind of bug report about the already found bugs and how the bugs should be corrected. My question is now. Will there be some kind of official patches for CHARMM22 or are we all left by our selves? Or is the CHARMM22 from Molecular Simulations the official patch? This has of cource a further price tag on it. Peace, -leif laaksonen --------------------------------------------------------- laaksone@csc.fi Centre for Scientific Computing P.O. Box 40 SF-02101 Espoo FINLAND Phone: 358 0 4572378 Telefax: 358 0 4572302 Voice Mail: 358 486257407 "In every job to be done there is an element of fun" Mary Poppins --------------------------------------------------------- -- From: Ed Moret Subject: 1:charges 2:overheating To: charmm@ruucmc.far.ruu.nl Date: Wed, 21 Oct 92 16:20:15 MET Name: Ed Moret Organisation: Utrecht University Phone: 030 - 536958 Mailer: Elm [revision: 66.25] Status: RO Dear CHARMm users, I just entered this bbs. Already I want to post two questions. If you decide that these items have been discussed at length before, please DO send your ideas using direct mail to me. We are using Polygens CHARMm 21.3, awaiting 22. 1. We want to simulate covalent drug binding to DNA. Unfortunately, this forces us to recalculate charges on the standard residues guanine etc. In order to use compatible charges for our drugs and for the new guanine etc. residues, we prefer to use the same methods as used in creating the current RTFs. HOW are these charges derived; The Nillson/Karplus article gives no info. Anticipating answers and presuming that we will not be able to use the same methods: what would be the best way to create new ones. - Quanta charge templates and charge smoothing - MNDO ESP derived charges - STO-3G/DMA charges I know that general agreement would favour the Electr. Pot. derived charges but will this be in the nature of the CHARMm program or disruptive? Please share your views with me. 2. Most often, in simulations with 30 angstrom water boxes and periodic boundary conditions (and SHAKE) our DNA hexamer simulations increase temperature to enormous figures. Rescaling the velocities does help, but prevents free simulations ofcourse (variable IEQFRQ). I am sure that we are not facing an intrinsic CHARMm PBC problem, so obviously we are doing something wrong. But what? Please respond to this one, because calculations at 1000 Kelvin are quite worthless for drug simulations, since not many people have a fever like this. Thanks for your time, Ed Moret From mitchell_t%frgen.dnet@smithkline.com Thu Oct 22 04:29:49 1992 Received: from phinet.smithkline.com by emperor.harvard.edu via SMTP (920110.SGI/911001.SGI) for dchin id AA22370; Thu, 22 Oct 92 04:29:49 -0400 Received: from FRGEN.DECnet MAIL11D_V3 by smithkline.com (5.57/Ultrix2.4-C) id AA22494; Thu, 22 Oct 92 04:04:49 -0400 Date: Thu, 22 Oct 92 04:04:48 -0400 Message-Id: <9210220804.AA22494@smithkline.com> From: mitchell_t%frgen.dnet@smithkline.com To: "charmm-bbs@emperor.harvard.edu"%INET.dnet@smithkline.com Subject: Partial charge calculations for use with CHARMm Status: RO A very quick way to calculate charges which "look" like the sort of numbers used in AMINO.RTF etc. is to use CNDO within Gaussian. The Gaussian version differs from the "normal" QCPE version as the wave-function is de-orthogonalised before the Mullikan analysis is performed. This results in charges which are more like ab-inito charges favoured by CHARMm (Don't ask me why - I'm not a Quantum Chemist!). I think I can dig out some references to the use of CNDO and de- orthogonalised wave functions if anyone would find them uselful. If no-one can come up with the definitive method for calculating charges, or this method is something silly like 6-31G**, how about comparing the charges generated by our "favourite" methods on a standard set of fragments (how about blocked Asp, Arg, Trp) with AMINOH.RTF ?. Tim Mitchell MITCHELL_T%FRGEN.DNET@SMITHKLINE.COM SmithKline Beecham Brockham Park Phone +737 36 4535 Betchworth Surrey RH3 7AJ, U.K. Fax +737 36 4539 From evanseck@tammy Thu Oct 22 08:31:03 1992 Received: from tammy.harvard.edu by emperor.harvard.edu via SMTP (920110.SGI/911001.SGI) for dchin id AA22454; Thu, 22 Oct 92 08:31:03 -0400 Received: from localhost by tammy.harvard.edu (5.64/10.0) id AA19434; Thu, 22 Oct 92 08:23:57 -0400 Message-Id: <9210221223.AA19434@tammy.harvard.edu> To: charmm-bbs Subject: Re: Partial charge calculations for use with CHARMm In-Reply-To: Your message of Thu, 22 Oct 92 04:04:48 -0400. <9210220804.AA22494@smithkline.com> Date: Thu, 22 Oct 92 08:23:55 -0400 From: evanseck@tammy Status: RO Dear Tim, You are right that there seems to be no definitive way to compute charges. If you are using semiempirical wavefunctions then you ought to try AM1 or PM3. I've had problems with CNDO in the past concerning the minimization of small organic structures, yet have never examined the resulting charges (which may be OK?). The AM1 (PM3) charges are close to those computed for moderate sized basis sets (6-31G*) using ab initio theory. In our parameter development for CHARMM we use 6-31G* to compute the initial charges on small organic prototypes then adjust them appropriately to achieve certain experimental observables (density, ...), which eventually become the charges on the standard set of fragments. There is a ton of literature already on this subject. I hope this helps, if not, then let me know. --Jeff Evanseck From mitchell_t%frgen.dnet@smithkline.com Thu Oct 22 10:05:45 1992 Received: from harvard.harvard.edu by emperor.harvard.edu via SMTP (920110.SGI/911001.SGI) for dchin id AA22493; Thu, 22 Oct 92 10:05:45 -0400 Received: by harvard.harvard.edu (5.54/a0.25) (for charmm-bbs@emperor) id AA23478; Thu, 22 Oct 92 10:05:00 EDT Received: from FRGEN.DECnet MAIL11D_V3 by smithkline.com (5.57/Ultrix2.4-C) id AA24632; Thu, 22 Oct 92 10:00:07 -0400 Date: Thu, 22 Oct 92 10:00:07 -0400 Message-Id: <9210221400.AA24632@smithkline.com> From: mitchell_t%frgen.dnet@smithkline.com To: charmm_bbs@smithkline.com Subject: Re: Partial charges / CHARMm Status: RO ----- Transcript of session follows ----- >>> RCPT To: <<< 550 ... User unknown 550 ... User unknown ----- Unsent message follows ----- Received: from FRGEN.DECnet MAIL11D_V3 by smithkline.com (5.57/Ultrix2.4-C) id AA24570; Thu, 22 Oct 92 09:48:17 -0400 Date: Thu, 22 Oct 92 09:48:16 -0400 Message-Id: <9210221348.AA24570@smithkline.com> From: mitchell_t%frgen.dnet@smithkline.com To: charmm_bbs@smithkline.com Subject: Re: Partial charges / CHARMm Jeff, I agree that geometry optimisation with CNDO is best avoided !. I am running a single point CNDO (from Gauss. 90) calculation on a structure optimised by a suitable method - such as AM1. Mulliken charges from "QCPE" AM1/PM3 can result in some strange mumbers, although AM1/PM3 potential fitted charges seem OK. My Quantum Chemistry colleagues tell me that there are problems using AM1/PM3 within Gaussian to calculate charges (with de-orthogonalised wavefunction) but I don't remember/understand what they are. Maybe I'll run some calculations on the fragments I suggested before and send them to the bbs. I don't have a machine big enough to run 6-31G* on these fragments - could you or anyone else out there help ?. Cheers, Tim Mitchell From Leif.Laaksone@csc.fi Fri Oct 23 09:16:21 1992 Received: from convex.csc.fi by emperor.harvard.edu via SMTP (920110.SGI/911001.SGI) for dchin id AA23147; Fri, 23 Oct 92 09:16:21 -0400 Received: from [128.214.46.163] (laaksone.csc.fi) by csc.fi with SMTP id AA13527 (5.65c8+/IDA-1.4.4 for charmm-bbs@emperor.harvard.edu); Fri, 23 Oct 1992 15:15:11 +0200 Date: Fri, 23 Oct 1992 15:15:11 +0200 Message-Id: <199210231315.AA13527@csc.fi> To: charmm-bbs From: Leif.Laaksone@csc.fi Subject: CHARMM22 benchmark Status: R If any kind soul would like to comment on my question about the FASTer option. I'm very surprised how slow the new CHARMM22 is. I made the same calculation with both the CHARMm21 (Polygen/MS) and CHARMM22 (Harvard). I did not use the FASTer option for CHARMm21 (in fact I never managed to use it successfully) but I used it for CHARMM22 and ended up using about the same amount of cpu-time. Does this mean that CHARMM22 is slower than CHARMm21 or did Polygen/MS modify the code to run faster? Can anybody tell me how well CHARMM/CHARMm is in general vectorized. The small benchmark I have used indicates that the program is not too much vectorized. Benchmark calculation with the new CHARMM22. The test is based on the 'solanal.inp' test case included in the original CHARMM22 tape. The reported time is the used cpu and system time in seconds. The memory on these machines has been between 32M and 256 M. ================================================================ Machine cpu time/system time (s) ================================================================ SGI Indigo Fast off: 564.5/2.1 on: 449.4/1.6 SGI 380VGX Fast off: 554.2/1.3 on: 465.2/1.5 IBM RS/6000 - 340 Fast off: 412.8/0.7 on: 266.9/0.9 Convex 220 Fast off: 527.3/15.9 on: 338.8/9.9 Fast vector: 338.5/14.2 Convex C3820 Fast off: 243.9/1.5 on: 127.1/1.2 Fast vector: 126.7/1.2 SUN SparcServer 390 Fast off: 1058.0/34.48 on: 841.9/22.48 SUN SparcServer 690MP Fast off: 575.0/21.13 on: 443.0/15.19 -leif laaksonen Ps. I just wish CHARMM could be that well supported and documented as Axel Brunger's X-PLOR is. He even sends out patches. === The benchmark numbers are not an official statement from the computer centre but reflects rather my own private activities. === --------------------------------------------------------- laaksone@csc.fi Centre for Scientific Computing P.O. Box 40 SF-02101 Espoo FINLAND Phone: 358 0 4572378 Telefax: 358 0 4572302 Voice Mail: 358 486257407 "In every job to be done there is an element of fun" Mary Poppins --------------------------------------------------------- From sfleisch@hydra.convex.com Fri Oct 23 11:38:27 1992 Received: from harvard.harvard.edu by emperor.harvard.edu via SMTP (920110.SGI/911001.SGI) for dchin id AA23239; Fri, 23 Oct 92 11:38:27 -0400 Received: by harvard.harvard.edu (5.54/a0.25) (for charmm-bbs@emperor) id AA20733; Fri, 23 Oct 92 11:37:42 EDT Received: from hydra.convex.com by convex.convex.com (5.64/1.35) id AA14627; Fri, 23 Oct 92 10:37:33 -0500 Received: by hydra.convex.com (5.64/1.28) id AA01899; Fri, 23 Oct 92 10:37:32 -0500 Date: Fri, 23 Oct 92 10:37:32 -0500 From: sfleisch@hydra.convex.com (Steve Fleischman) Message-Id: <9210231537.AA01899@hydra.convex.com> To: Leif.Laaksone@csc.fi, charmm-bbs%emperor@harvard Subject: Re: CHARMM22 benchmark Status: R I would suggest that you submit the actual input file so that others can check this out. Steve From polygen!asteroid!ryszard Sat Oct 24 09:03:24 1992 Received: from BU.EDU by emperor.harvard.edu via SMTP (920110.SGI/911001.SGI) for dchin id AA24762; Sat, 24 Oct 92 09:03:24 -0400 Received: from polygen.UUCP by BU.EDU (1.99) Sat, 24 Oct 92 09:02:35 -0400 Received: from asteroid.polygen by polygen.com (4.0/SMI-4.0) id AA06737; Sat, 24 Oct 92 00:39:12 EDT Received: by asteroid.polygen (4.1/SMI-4.1) id AA23867; Fri, 23 Oct 92 23:44:51 CDT Date: Fri, 23 Oct 92 23:44:51 CDT From: polygen!asteroid!ryszard (Ryszard Czerminski X 285) Message-Id: <9210240444.AA23867@asteroid.polygen> To: charmm-bbs@bu.UUCP Status: R Some comments to old and more recent charmm-bbs messages: I repeat short messages or some keywords from longer messages... ============================================================= Does anyone out there have programs to read DCD and CSR files generated by Quanta ? Vikram Varma Simon Fraser University I have a program called DA which can read and analyze charmm DCD files and does some others useful things. At least useful for me some time ago. Good news is that it is for free and simple (relatively) to modify and taylor to your own needs, bad news is that is not very well documented and I cannot promise any support. If anybody is interested - let me know. ============================================================= From: "Mathai Mammen" Subject: Rotational Partition Function and Solvent Well, this is good question and from the way I am starting my comment you can guess that I do not have any good answer. My intuition goes however different way i.e. it says that considering rotational partition function should be more important for small molecules than for e.g. big protein in solvent which has rotation severely hindered if not completely frozen. As far as difference between gas phase and solvent goes I think there should be significant difference since solvent is restricting rotation and probably proper way to take it into account would be to consider some kind of hindered rotation with randomly changing barrier. How it would affect partition function ? I do not know. On the other hand if you are using MD to calculate some properties (from the fact that this question was posted on charmm-bbs I deduce that you are using charmm!) you should realize that you are using accurate method and you do not have to consider some inferior approximations like harmonic oscillator an rigid rotor. Well, method is of course accurate in principle and this is because of limitations of computer hardware i.e. we are limited to relatively short trajectories and probing of the phase space is highly incomplete. === From: Ed Moret Subject: 1:charges 2:overheating Since there were already several comments on subject: 1:charges I think I can add one more: In general using Mulliken charges is not a very good idea since they even do not reproduce dipol moments ! As Jeff Evanseck pointed out there is a lot of literature on the subject and if anybody is interested I can dig out some references. Better way is to include atomic and interatomic charges and higher moments (dipols, quadrupols...) or if we are restricted for practical reason to atomic charges to use charges fitted to reproduce electrostatic potentials (as Ed Moret pointed out) from some reasonable wave function. As far as consistency issue with charmm charges is concerned you pobably will be better of if you go along the path desrcibed by Jeff. That's all for today, Ryszard Czerminski, MSI From moret@far.ruu.nl Sat Oct 24 15:29:33 1992 Received: from accucx.cc.ruu.nl by emperor.harvard.edu via SMTP (920110.SGI/911001.SGI) for dchin id AA24904; Sat, 24 Oct 92 15:29:33 -0400 Received: from ruucmc.far.ruu.nlfar.ruu.nl (ruucmc.far.ruu.nl) by accucx.cc.ruu.nl with SMTP id AA29161 (5.65c/IDA-1.4.4 for ); Sat, 24 Oct 1992 20:28:48 +0100 Message-Id: <199210241928.AA29161@accucx.cc.ruu.nl> Received: by ruucmc.far.ruu.nl (16.6/16.2) id AA16073; Sat, 24 Oct 92 20:26:37 +0100 From: Ed Moret Subject: Answers to 1.charges,2.overheating To: charmm@ruucmc.far.ruu.nl Date: Sat, 24 Oct 92 20:26:36 MET Name: Ed Moret Organisation: Utrecht University Phone: 030 - 536958 Mailer: Elm [revision: 66.25] Status: R Dear CHARMm-users, The question of what charges to recalculate for CHARMm was discussed by Tim Mitchell, Jeff Evanseck and Ryszard Czerminski on the bbs. The fragment RTF charges are 6-31G* derived and adjusted to fit experiment. For users, it seems wise to calculate potential fitted atomic charges. STO-3G/DMA gives results comparable to the Amber charges (we tested only two DNA bases), but MNDO/ESP charges may be comparable or better. (I think Bessler et al. pointed out that for ESP calculations, MNDO is to be preferred over AM1 and PM3). The overheating problems are history. Initially, I thought it not too important how far the structure was minimised before dynamics. It would be loss of information to start the simulation in a conformational minimum too deep to be escaped. However, after adding a box of water and using periodic boundary conditions, it may be necessary to minimise for a long time (10000 steps ABNR in our case). Scott Flicker and David Busath are acknowledged for comments. Anyway, our big mistake was most likely the use of SHAKE bond constraints during minimisation. Even though SHAKE is applied during dynamics, it seems nonsense to do the same during minimisation. Thanks for your time, Ed Moret -- ------------------------------------------------------------------------- E.E. Moret @more@ Computational Medicinal Chemistry/Department of Pharmaceutical Chemistry Faculty of Pharmacy/Utrecht University/the Netherlands Mail Address P.O.Box 80082 3508 TB Utrecht Telephone (31-30)536979/536958 Telefax (31-30)516674 E-mail moret@far.ruu.nl, moret@cc.ruu.nl ------------------------------------------------------------------------- From mitchell_t%frgen.dnet@smithkline.com Mon Oct 26 11:52:36 1992 Received: from phinet.smithkline.com by emperor.harvard.edu via SMTP (920110.SGI/911001.SGI) for dchin id AA26304; Mon, 26 Oct 92 11:52:36 -0500 Received: from FRGEN.DECnet MAIL11D_V3 by smithkline.com (5.57/Ultrix2.4-C) id AA24946; Mon, 26 Oct 92 11:46:25 -0500 Date: Mon, 26 Oct 92 11:46:24 -0500 Message-Id: <9210261646.AA24946@smithkline.com> From: mitchell_t%frgen.dnet@smithkline.com To: "charmm-bbs@emperor.harvard.edu"%INET.dnet@smithkline.com Subject: Re: Charges for CHARMm Status: R Jeff, I agree that geometry optimisation with CNDO is best avoided !. I am running a single point CNDO (from Gauss. 90) calculation on a structure optimised by a suitable method - such as AM1. Mulliken charges from "QCPE" AM1/PM3 can result in some strange mumbers, although AM1/PM3 potential fitted charges seem OK. My Quantum Chemistry colleagues tell me that there are problems using AM1/PM3 within Gaussian to calculate charges (with de-orthogonalised wavefunction) but I don't remember/understand what they are. Maybe I'll run some calculations on the fragments I suggested before and send them to the bbs. I don't have a machine big enough to run 6-31G* on these fragments - could you or anyone else out there help ?. Cheers, Tim Mitchell ------------------------------------------------------------------------------- | _____ ____ || SmithKline Beecham Pharmaceuticals Ltd | | / ___\\\\ _ \ || Brockham Park | | | |___ \\|_| | || Tim Mitchell (Comp. Chem.) Betchworth | | \____ \ | _ < || Phone (0)737 36 4535 Surrey | | ____| | | |_| | || Fax. (0)737 36 4539 RH3 7AJ | | \_____/ |____/ || mitchell_t%frgen.dnet@smithkline.com U.K. | ------------------------------------------------------------------------------- From polygen!asteroid!ryszard Wed Oct 28 09:03:14 1992 Received: from BU.EDU by emperor.harvard.edu via SMTP (920110.SGI/911001.SGI) for dchin id AA29155; Wed, 28 Oct 92 09:03:14 -0500 Received: from polygen.UUCP by BU.EDU (1.99) Wed, 28 Oct 92 09:02:26 -0500 Received: from asteroid.polygen by polygen.com (4.0/SMI-4.0) id AA09379; Tue, 27 Oct 92 14:44:16 EST Received: by asteroid.polygen (4.1/SMI-4.1) id AA06748; Tue, 27 Oct 92 13:49:58 CST Date: Tue, 27 Oct 92 13:49:58 CST From: polygen!asteroid!ryszard (Ryszard Czerminski X 285) Message-Id: <9210271949.AA06748@asteroid.polygen> To: charmm-bbs@bu.UUCP Status: RO Few days ago Ed Moret wrote ... Anyway, our big mistake was most likely the use of SHAKE bond constraints during minimisation. Even though SHAKE is applied during dynamics, it seems nonsense to do the same during minimisation. ... I do not see anything inherently wrong with using SHAKE during minimization as long as you are using "SHAKE PARAM..." option which should be anyhow used for most of the regular applications. This keeps shaken bond lengths at equilibrium values from parameter set (which is probably what you want), if you do not use PARAM then shaken bond lengths are kept at initial value (which probably you do not want). Ryszard Czerminski, MSI **Note ** I am posting this for: Eldbjoerg Heimstad Protein Crystallography Group University Of Tromsoe 9000 Tromsoe NORWAY email: eldbjorg@trypsin.chem.uit.no Please respond to charmm-bbs@emperor.harvard.edu Thanks, Donovan ** End Note ** -- Begin -- I have recently started to use CHARMm standalone, and have some problems which may seem trivial for others. I have old .DCD files from molecular dynamics simulations using QUANTA3.2/CHARMm21. From release notes on QUANTA3.3/CHARMm22 I understand that .DCD files have changed format. Am I able to format the old .DCD files with CHARMm22? commandfile: * Script to read parameter, psf, and ic files * reset open read unit 21 card name $CHM_DATA/MASSES.RTF read rtf unit 21 card close unit 21 open read unit 20 card name "laksivab.prm" read param unit 20 card close unit 20 open read unit 20 card name "laksivab.psf" read psf unit 20 card close unit 20 open read unit 20 card name "laksivab.ic" ic read unit 20 card close unit 20 READ SEQU CARD *leser sekvens * 242 ILE VAL GLY GLY TYR GLU CYS LYS ALA ASN SER GLN ALA TYR GLN VAL SER LEU ASN SER GLY TYR HIS PHE CYS GLY GLY SER LEU VAL ASN ASN THR TRP VAL VAL SER ALA ALA HIS CYS TYR LYS SER GLY ILE GLN VAL ARG LEU GLY GLU HIS ASN ILE ALA VAL THR GLU GLY ASN GLU GLN PHE ILE GLY SER SER LYS VAL ILE MET HIS PRO ASN TYR SER GLY GLY ASN LEU ASP ASN ASP ILE MET LEU ILE LYS LEU SER LYS PRO ALA THR LEU ASN SER TYR VAL GLN PRO VAL ALA LEU PRO SER SER CYS ALA PRO SER GLY THR MET CYS ILE VAL SER GLY TRP GLY ASN LEU SER GLY SER SER SER ALA ASP SER ASP THR LEU GLN CYS LEU ASP ILE PRO ILE LEU SER PHE SER SER CYS ASN ASP THR TYR PRO GLY GLN ILE THR ALA ALA MET PHE CYS ALA GLY TYR MET ALA GLY GLY LYS ASP SER CYS SER GLY ASP SER GLY GLY PRO VAL VAL CYS ASP GLY GLU LEU GLN GLY VAL VAL SER TRP GLY TYR GLY CYS ALA GLN PRO GLY ASP PRO GLY VAL TYR SER LYS VAL CYS ILE PHE SER GLY TRP ILE SER SER THR MET GLY CA BEN HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH !GENE MAIN SETU OPEN READ UNIT 18 CARD NAME "laksivab.CRD" COOR READ CARD UNIT 18 ! Convert binary coordinates into machine ! independent form OPEN READ UNIT 41 FILE NAME "simul.DCD" OPEN WRITE UNIT 61 CARD NAME dyna.dhx DYNA FORM FIRST 41 NUNIT 1 OUTPUT 61 STOP Note: I try to run with GENE MAIN SETU, but it did not work; should it work? LOG-file: here is the last part: **** Message from SEQRDR ***** THE SYSTEM CONTAINS 36 TITRATABLE GROUPS THE USER MUST PREDETERMINE THE PROTONATION STATE THROUGH THE SEQUENCE AND RTF HIS - 4 ASP - 10 GLU - 5 LYS - 7 TYR - 10 CHARMM> CHARMM> CHARMM> !GENE MAIN SETU CHARMM> CHARMM> OPEN READ UNIT 18 CARD NAME "laksivab.CRD" VOPEN> Attempting to open::laksivab.CRD:: OPNLGU> Unit 18 opened for READONLY access to laksivab.CRD CHARMM> COOR READ CARD UNIT 18 SPATIAL COORDINATES BEING READ FROM UNIT 18 TITLE> *EXPORT TO CHARMM CRD FROM LAKSIVAB.MSF WITH Q3.3 TITLE> * CHARMM> CHARMM> ! Convert binary coordinates into machine CHARMM> ! independent form CHARMM> CHARMM> OPEN READ UNIT 41 FILE NAME "simul.DCD" VOPEN> Attempting to open::simul.DCD:: OPNLGU> Unit 41 opened for READONLY access to simul.DCD CHARMM> OPEN WRITE UNIT 61 CARD NAME dyna.dhx VOPEN> Attempting to open::dyna.dhx:: OPNLGU> Unit 61 opened for WRITE access to dyna.dhx CHARMM> DYNA FORM FIRST 41 NUNIT 1 OUTPUT 61 READING TRAJECTORY FROM UNIT 41 NUMBER OF COORDINATE SETS IN FILE: 100 NUMBER OF PREVIOUS DYNAMICS STEPS: 40480 FREQUENCY FOR SAVING COORDINATES: 500 NUMBER OF STEPS FOR CREATION RUN: 50000 TITLE> * COORDINATES AFTER EQUILIBRATION DYNAMICS TITLE> * DATE: 12/27/91 12: 3:15 CREATED BY USER: eldbjorg TITLE> * ***** WARNING ***** SKIP= 1 WAS NOT A MULTIPLE OF THE FILE INTERVAL= 500 IT HAS BEEN RESET TO 500 ***** WARNING ***** BEGIN= 0 WAS NOT IN THE INPUT FILE IT HAS BEEN RESET TO 40480 ***** WARNING ***** BEGIN= 40480 WAS NOT A MULTIPLE OF SKIP IT HAS BEEN RESET TO 40500 start: end of file -1 apparent state: unit 41 named simul.DCD lately reading sequential unformatted external IO By the way, does it exist any documentation for CHARMm errors? Regards, Eldbjoerg Heimstad Protein Crystallography Group University Of Tromsoe 9000 Tromsoe NORWAY From Leif.Laaksone@csc.fi Wed Nov 11 01:10:50 1992 Received: from convex.csc.fi by emperor.harvard.edu via SMTP (920110.SGI/911001.SGI) for dchin id AA10198; Wed, 11 Nov 92 01:10:50 -0500 Received: from [128.214.46.163] (laaksone.csc.fi) by csc.fi with SMTP id AA07430 (5.65c8+/IDA-1.4.4 for charmm-bbs@emperor.harvard.edu); Wed, 11 Nov 1992 08:09:14 +0200 Date: Wed, 11 Nov 1992 08:09:14 +0200 Message-Id: <199211110609.AA07430@csc.fi> To: charmm-bbs From: Leif.Laaksone@csc.fi Subject: Re: CHARMm22 DCD files. Status: R According to the QUANTA User's Guide Supplement (page 155) there should be no problems for CHARMm22 to read old CHARMm21 produced dcd files. The only new thing is the symmetric shape index data record (float XTLABC(6)) and this record should not be touched in other cases. But who knows, perhaps somebody has introduced a programming error. My suggestion in any case is that you write a small piece of fortran code to format your dcd files. I don't have the code available but I have some code for doing the merging of dcd files. This code can easily be changed to format your files. Cheers, -leif laaksonen >Eldbjoerg Heimstad > >Protein Crystallography Group >University Of Tromsoe >9000 Tromsoe >NORWAY > >email: eldbjorg@trypsin.chem.uit.no > >-- Begin -- >I have recently started to use CHARMm standalone, and have some >problems which may seem trivial for others. >I have old .DCD files from molecular dynamics simulations using >QUANTA3.2/CHARMm21. From release notes on QUANTA3.3/CHARMm22 >I understand that .DCD files have changed format. >Am I able to format the old .DCD files with CHARMm22? > --------------------------------------------------------- laaksone@csc.fi Centre for Scientific Computing P.O. Box 40 SF-02101 Espoo FINLAND Phone: 358 0 4572378 Telefax: 358 0 4572302 Voice Mail: 358 486257407 "In every job to be done there is an element of fun" Mary Poppins --------------------------------------------------------- From eldbjorg@chem.uit.no Fri Dec 4 09:16:16 1992 Received: from benoni.Uit.No by emperor.harvard.edu via SMTP (920110.SGI/911001.SGI) for dchin id AA01253; Fri, 4 Dec 92 09:16:16 -0500 Received: from benoni by ppenoni.uit.no with SMTP (PP) id <00227-0@ppenoni.uit.no>; Fri, 4 Dec 1992 15:14:56 +0000 Received: from trypsin.Chem.Uit.No by benoni.uit.no (5.65+IDA/Babel-1.15/ABaa-1.2/Ultrix) id AAbenoni00223; Fri, 4 Dec 1992 15:14:51 +0100 Received: by trypsin.chem.uit.no (911016.SGI/ABaa-2.0mini) id AA22170; Sat, 5 Dec 92 00:15:33 +0100 Date: Sat, 5 Dec 92 00:15:33 +0100 From: eldbjorg@chem.uit.no Message-Id: <9212042315.AA22170@trypsin.chem.uit.no> To: charmm-bbs Subject: sample stream file Status: RO To David Busath: I am sorry for bothering you again about the copy of Charmm stream file, only do this if you have time to it. I have read dynamc.doc , so you really have to read one coordinate set each time with TRAJ READ? -- no simpler way to do it ? -- The DCD file contains 100 coordinate sets,so .............. Another question about N-terminal and C-terminal: I understand that AMINO.RTF (default topology file in QUANTA) contains default patching of NTER and CTER. I had to give in another oxygen atom at the cter end, because the C-atom bond to only one oxygen became unknown in Mol. Editor--- should it be necessary to explicitly have another oxygen atom (to get COO-) at CTER in PDB file ? Regards, Eldbjoerg P.S. Who runs QUANTA/CHARMm course next year? ####################################################### # Eldbjoerg Sofie Heimstad # # Protein Crystallography Group # # University Of Tromsoe, IMR # # 9037 Tromsoe, NORWAY phone: +47-83-45706 # # 44737 # # E-mail: eldbjorg@trypsin.chem.uit.no # ####################################################### Has anyone got a charmm with the contsant pressure code for hexagonal crystal types - the present version seems to deal with orthorhombic and cubic only. thanx chandra CHARMM22 version should support CUBIC, ORTHOROMBIC and TRICLINIC crystal types. Ryszard Czerminski, MSI