From owner-hyperchem Sat Apr 1 15:02:50 1995 Date: Sat, 1 Apr 1995 14:39:28 -0300 (BST) From: Edecio C Neto Subject: visualization of PDB files To: hyperchem@hyper.com I have been trying to open PDB files downloaded from the PDB database via }ithe RETRIEVE NCBI/NIH server. Every single time I tried it I got a message error like "missing parameter at line 283" or the like. Should I modify the file before opening it at Hyperchem? Of course I deleted the E-mail and query text headers, leaving only the PDB file text as is. Edecio Cunha-Neto, M.D., Ph.D. ________________________________________________________________________ From owner-hyperchem Mon Apr 3 13:16:27 1995 From: "Jens Spanget-Larsen" Organization: Roskilde Universitetscenter To: hyperchem-request@hyper.com, hyperchem@hyper.com Date: Mon, 3 Apr 1995 12:46:28 +0100 Subject: IR Intensities Dear HyperChem'ers, I would like to know, how the theoretical quantities "Integrated Infrared Band Intensities in km/mol" listed in the HyperChem log-file are computed. What is the relation between these quantities and the calculated vibrational "Frequencies" and "Derivatives of Dipole Moments..(Debye/Angstrom/AMU)"? Jens >--< _________________________________ _________________________________ | | | | Jens Spanget-Larsen | Phone: +45 46757711 | | Associate Professor, dr.scient. | - direct: +45 46757781 * 2710 | | Department of Chemistry (17.2) | - private: +45 42840320 | | Roskilde University | Fax: +45 46757721 [Dept.] | | RUC, P.O. Box 260 | - +45 46757401 [RUC] | | DK-4000 Roskilde, Denmark | Internet: JSL@mmf.ruc.dk | |_________________________________|_________________________________| ________________________________________________________________________ From owner-hyperchem Mon Apr 3 14:18:23 1995 From: "Slawomir Berski" To: hyperchem@hyper.com Date: Mon, 3 Apr 1995 14:06:52 GMT+1 Subject: Li-problems Hi, I would like to perform some calculations on organolithium compounds using HyperChem software. Unfortunately there is a lack of lithium parameters for PM3 method in this program ver. 3.0. I tried to adopt parameters prepared by E.Anders , R.Koch and P.Freunschaft (J.Comp.Chem 14(1993)1301) designed for MOPAC 6/PC program but HyperChem requires additional parameters for monopole- monopole, dipole-dipole and quadrupole-quadrupole interactions with symbols: AM,AD and AQ at positions: 12,13,14 in pm3_2.apb file. Could anyone give me advise how to evaluate these parameters or maybe where can I find such parameters for lithium atom but designed for HyperChem program. Sincerely Slawek ________________________________________________________________________ From owner-hyperchem Mon Apr 3 17:59:32 1995 Date: Mon, 3 Apr 95 10:17:18 -0400 From: polowin@hyper.hyper.com (Joel Polowin) To: hyperchem@www.hyper.com Subject: Re: ab initio for HyperChem > From: "Dr. Steven Desjardins" > Date: Tue, 28 Mar 1995 12:10:56 EDT > > Is it true that a new release of hyperchem is available with ab > initio methods? Is this in the PC version, and, if so, what are the > details, e.g., price, methods, basis sets, etc.? I apologise for the delay in responding to your note. After I made the error of destroying confidentiality by releasing information in response to an earlier query, I was asked to wait until after the formal announcement at the ACS meeting yesterday before providing further details. Here is the information file that we have compiled; if you have further questions or feel that something here requires clarification, amplification or correction, please let me know. HyperChem 4.5 - What's New -------------------------- The most significant change between versions 4 and 4.5 of HyperChem is the addition of ab initio functionality. Using any basis set of s, p, or d orbitals, ab initio calculations of the following kinds can be performed: * SCF calculations of single points * geometry optimizations * molecular dynamics * vibrational analysis * optical spectroscopy Basis Sets ---------- A large number of basis sets are supplied, and users are also able to define their own basis sets. HyperChem uses a simple text file format to describe basis sets, making users' modifications easy. Considerable flexibility in the choice of a basis set is available, including the ability to simply select a set of atoms with the mouse and assign a basis set to those atoms only. This makes it possible to have different basis sets assigned to different atoms, or even to select an atom and interactively assign a primitive s, p, sp, d, or spd shell to that atom with the user specifying an exponent for the shell. It is even possible to interactively select a set of atoms and have their nuclei treated as "ghost nuclei" - HyperChem uses the basis functions but neglects the nuclear interactions. Calculations ------------ Single-point correlation energy calculations at the MP2 level can also be performed. The ab initio calculations are completely integrated into HyperChem's usual graphical support, allowing users to visualize orbital energy plots, electron densities, IR and UV spectra, etc. Chemical reactions can be explored by performing molecular dynamics with forces computed by ab initio methods. And More! --------- Commands have been added to the scripting language to allow all of the added functions to be run by scripts, or under the control of external software such as Visual Basic or Microsoft Excel via dynamic data exchange (DDE). As usual, Hypercube supplies documentation which not only shows users how to operate the software, but also provides a solid introduction to the theory of ab initio calculations. This makes HyperChem an extremely useful educational tool. Release 4.5 of HyperChem also allows a user to explore isotope effects in vibrational analysis and input ChemDraw files. For comparison with other platforms and programs, a minimal basis set wave function calculation for a glycine dipeptide zwitterion takes about 3 minutes on a 90 MHz Pentium computer. We expect to ship Release 4.5 for Windows in June 1995, and Release 4.5 for SGI in July 1995. As far as pricing goes, suggested retail prices for the Windows version as of April 1 will be as follows, given in $US, commercial (acedemic): Release 4.5 for Windows, standalone: 2,595 (1,295) network, 1 license 2,595 (1,295) 5 license 10k (5k) 10 license 18k (9k) 25 license 30k (15k) 50 license 40k (20k) unlimited 60k (30k) Extra rel. 4.5 for Windows manual set 350 (175) Release 4 to 4.5 update 795 (495) 3 4.5 1,095 (650) 2 4.5 1,595 (895) 3 to 4 update 395 (195) 2 4 995 (495) Update of N non-networking to N-license network version of v. 4.5: free The commercial price of a one-license Release 4.5 of HyperChem for SGI will be US$2995, while the government/academic price will be $1495. Volume discounts and site licenses are available; we expect to implement an upgrade policy for users of earlier SGI packages but the details have not yet been settled. ------------ Joel Polowin, Ph.D. Manager, Scientific Support Email to: polowin@hyper.com Hypercube Inc, 7-419 Phillip St, Waterloo, Ont, Canada N2L 3X2 (519)725-4040 Info requests to: info@hyper.com Support questions to: support@hyper.com Email group: Send "subscribe hyperchem" to hyperchem-request@hyper.com ________________________________________________________________________ From owner-hyperchem Mon Apr 3 19:35:14 1995 Date: Mon, 3 Apr 95 13:55:00 -0400 From: polowin@hyper.hyper.com (Joel Polowin) To: hyperchem@hyper.com Subject: Re: Error Messages > Date: Fri, 31 Mar 1995 09:18:01 -0600 (CST) > From: Jerry Stacy Tucker > > I have a question concerning an error message I received after solvating > a protein. I created a periodic box about the protein and initiated my > calculations. In a few moments, I received an error that "two or more > atoms are in the same position." > > Is there a way to correct this problem? Since the error message did not > appear until after I began using water molecules in my calculations, I > believe they're the cause of the error. Can I simply find the offending > water molecule and remove it? Hmm. "It's not supposed to do THAT." If you can spot the water molecule and delete it, that should solve the problem -- if that is the cause. When you add a periodic box of water molecules to a structure, HyperChem uses a database file of water molecules and their locations; it repeats the total structure as many times as necessary to give the size of box you need full of water. And it's supposed to delete any water molecule that would come within the specified distance of the existing structure, to avoid exactly the sort of problem that you're seeing here. All that I can think of is that somehow the periodicity of the boxes of water molecules, placed side by side, is allowing an atom at one side to project to its duplicate image on the other side of the box. Another possibility would be if you had merged in a structure several times, after setting "translate-merged-systems" off with a script command. I'm trying to think of some way to check the structure for overlapping identical atoms. All that I can think of off-hand would be to take the structure's file in .HIN format and sort the lines in the file, to try to find duplicate lines. This would be easier in UNIX than in DOS, since there are a couple of UNIX utilities that will not only sort files but also warn of duplicated lines. You might be able to read a .HIN file into a good spreadsheet program, splitting the data up into spreadsheet columns; this would again let you sort the atoms by their coordinates, and perhaps pick up duplicates. Regards, Joel ------------ Joel Polowin, Ph.D. Manager, Scientific Support Email to: polowin@hyper.com Hypercube Inc, 7-419 Phillip St, Waterloo, Ont, Canada N2L 3X2 (519)725-4040 Info requests to: info@hyper.com Support questions to: support@hyper.com Email group: Send "subscribe hyperchem" to hyperchem-request@hyper.com ________________________________________________________________________ From owner-hyperchem Mon Apr 3 20:02:47 1995 Date: Mon, 3 Apr 95 14:35:11 -0400 From: polowin@hyper.hyper.com (Joel Polowin) To: hyperchem@hyper.com Subject: Re: macrocycles and internal rotation > Date: Fri, 31 Mar 1995 17:58:39 +0000 (GMT) > From: KANEPCHEM <94970459@vax1.dcu.ie> > > I am using molecular mechanics to study the structure of calixarenes, a class > of macrocycle. Calixarenes consist of 4 to 14 benzene rings, joined by -CH2- > bridges to form one big ring. > > I wish to know how to rotate a benzene ring andany groups attached through > an angle of for example 20 degrees, about the axis containing the atoms in the > benzene ring where the which are bonded to the -CH2- groups. > > Currently, I am trying to do this by attempting to create a plane containing > the aforementioned two atoms which would make an angle of 10 degrees with the > benzene ring and then I could REFLECT the NAMED SELECTION of the benzene ring > and the attached groups through this plane. However I don't know how to create > such a plane. The proper procedure is as follows: 1. Select the two atoms that define the axis you wish to rotate about. 2. Name the selection as LINE, using Select/Name Selection. 3. Select the structure that you wish to rotate about that axis. HyperChem will not allow one to rotate only part of a structure because of the danger of causing serious distortions of the bonds between the selected and unselected parts. Therefore: 4. Delete the bonds that join the selected part to the rest of the structure. Be careful to delete the bonds and not the atoms -- as necessary, use zooming (or the key) to let you right-click on the bond without touching an atom with the cursor. 5. Use Edit/Rotate; specify "Rotate About LINE", the angle, and that the rotation should Apply To Molecules. 6. Recreate the bonds that you had to break. Since you've been rotating about the axis defined by two of the atoms that define those bonds, those atoms will not have moved; the recreated bonds will be the same as the ones you broke. Regards, Joel ------------ Joel Polowin, Ph.D. Manager, Scientific Support Email to: polowin@hyper.com Hypercube Inc, 7-419 Phillip St, Waterloo, Ont, Canada N2L 3X2 (519)725-4040 Info requests to: info@hyper.com Support questions to: support@hyper.com Email group: Send "subscribe hyperchem" to hyperchem-request@hyper.com ________________________________________________________________________ From owner-hyperchem Mon Apr 3 20:55:36 1995 Date: Mon, 3 Apr 95 15:22:40 -0400 From: polowin@hyper.hyper.com (Joel Polowin) To: hyperchem@hyper.com Subject: Re: visualization of PDB files > Date: Sat, 1 Apr 1995 14:39:28 -0300 (BST) > From: Edecio C Neto > > I have been trying to open PDB files downloaded from the PDB database via > }ithe RETRIEVE NCBI/NIH server. Every single time I tried it I got a > message error like "missing parameter at line 283" or the like. Should I > modify the file before opening it at Hyperchem? Of course I deleted the > E-mail and query text headers, leaving only the PDB file text as is. You should not need to modify the PDB file before reading it into HyperChem. It is possible that you had a problem with ASCII/binary file transfer at some point in the downloading and file transfers, though it does not sound much like that kind of error. If the PDB file contains a non-standard residue, it will give a message like "Modified or unknown residues: CPT 25 CPT 26 CPT 27 ...", where CPT is the name of a non-standard residue (in this case, cisplatin) as given in the file. In this case it may be necessary to add the new residue(s) to the CHEM.TPL file, as described in Appendix E of the Reference Manual. If the file is not completely in the proper PBD format, you may need to use the script command "non-standard-pdb-names no". This is, I believe, of particular relevance if the atom names are left-justified instead of right-justified. If these suggestions still do not help you solve the problem, if you will E-mail me a copy of the file, or tell me what the PDB file is called so that I can look for it myself, I will look at the problem here. Regards, Joel ------------ Joel Polowin, Ph.D. Manager, Scientific Support Email to: polowin@hyper.com Hypercube Inc, 7-419 Phillip St, Waterloo, Ont, Canada N2L 3X2 (519)725-4040 Info requests to: info@hyper.com Support questions to: support@hyper.com Email group: Send "subscribe hyperchem" to hyperchem-request@hyper.com ________________________________________________________________________ From owner-hyperchem Mon Apr 3 21:35:25 1995 Date: Mon, 3 Apr 1995 15:47:52 -0400 From: Yufie Guo To: hyperchem-request@hyper.com, hyperchem@hyper.com, JSL@mmf.ruc.dk Subject: Re: IR Intensities Cc: guo, polowin >Dear HyperChem'ers, >I would like to know, how the theoretical quantities "Integrated >Infrared Band Intensities in km/mol" listed in the HyperChem log-file >are computed. What is the relation between these quantities and the >calculated vibrational "Frequencies" and "Derivatives of Dipole >Moments..(Debye/Angstrom/AMU)"? >Jens >--< The relationship between the Integrated Infrared Band Intensity (in units of km/mol), I(B), and the IR intensity (in unit of D^2 u^-1 A^-2: D=Debye, A=Angstrom, and u=atomic mass unit), I is pi * Na * g I(B) = ---------------------- I 1000 C**2 * 2.302581 where pi = 3.14159.... Na = Avogadro's constant. g = Degeneracy factor. C = Speed of light. For details, please look at the reference: B.A.Hess,Jr. and L.J.Schaad, Chem. Rev. 86 (1986), 709-730. Hope this helps! Yufei Guo *************************************************************************** * Yufei Guo, Ph.D. (guo@hyper.com) * * Senior Scientist, * * Hypercube Inc, * * 419 Phillip St, Waterloo, Ont, Canada N2L 3X2 (519)725-4040 * *************************************************************************** ________________________________________________________________________ From owner-hyperchem Tue Apr 4 14:25:26 1995 Date: Tue, 4 Apr 1995 16:55:21 UTC+0100 From: "Jose G. de la Campa" Subject: about reactions To: hyperchem@hyper.com In the frequent asked questions there is one about how can I simulate a reaction?. The answer says the best method is the use of set velocity and molecular dynamics by using a semi empirical QM method. This seem to work properly with the simple example you propose, (and even in that case we have been forced to change the velocity to obtain good results). However, when we try other reactions we usually get very bad results (I should say horrible results). Is there any way to optimize the initial conditions to obtain "normal" results. Are there any rules to avoid wasting time with useless results. Is there any way to avoid the rebound of the attacking molecule when it approaches to much to the second molecule? Has anybody used this method to simulate reactions? Is this supposed to be similar to the method used in MOPAC in which you successively approach two molecules by forcing the distance between them to be proggressively shorter and you calculate the minimum energy conformation for every case? Any help will be appreciated. Jose Gonzalez de la Campa Instituto de Ciencia y Tecnologia de Polimeros, Madrid, Spain ________________________________________________________________________ From owner-hyperchem Wed Apr 5 15:08:05 1995 Date: Wed, 5 Apr 95 11:15:05 -0400 From: polowin@hyper.hyper.com (Joel Polowin) To: "Jose G. de la Campa" Subject: Re: about reactions Cc: hyperchem@www.hyper.com > Date: Tue, 4 Apr 1995 16:55:21 UTC+0100 > From: "Jose G. de la Campa" > In the frequent asked questions there is one about how can I simulate a > reaction?. The answer says the best methos is the use of set velocity and > molecular dynamics by using a semi empirical QM method. This seem to work > properly with the simple example you > propose, (and even in that case we have been forced to change the velocity > to obtain good results). However, when we try other reactions we usually > get very bad results (I should say horrible results). Is there any way to > optimize the initial conditions to obtain "normal" results. Are there any > rules to avoid wasting time with useles results. Is there any way to avoid > the rebound of the attacking molecule whent it approaches to much to the > second molecule? Has anybody used this method to simulate reactions? Is > this supposed to be similar to the method used in MOPAC in which you > successively approach two molecules by forcing the distance between them > to be proggressively shorter and you calculate the minimum energy > conformation for every case? Any help will be appreciated. No; this method is much more like a classical trajectory simulation. By following this method, you are not forcing the structures closer together and minimizing at each step; you are specifying that the structures are approaching each other with an initial velocity defined by the atomic vectors, and letting nature take its course. The kinetic energy of the system exchanges with potential energy. If the kinetic energy overcomes the potential barrier to reaction, then reaction occurs. All that is happening when the "attacking molecule" rebounds is that for that particular approach, it does not have enough energy to overcome the reaction barrier. To solve the problem, you would specify either a higher approach velocity, or a different conformation with a lower activation barrier. For example, in the reaction F + H2 --> HF + H, if the F approaches along the H--H axis, the barrier to reaction is only a few kcal/mol. If it is moving in the plane normal to that axis, to strike at the mid-point of the H--H bond, the barrier is *much* higher. An F atom that was moving quickly enough to react easily in the collinear case would almost certainly just bounce off the midpoint of the H2 molecule. In the more usual trajectory calculations, the situation is complicated by the fact that the molecule being approached by the single atom would be rotating and vibrating. Those energy factors also contribute -- in the above case, a slow F atom might still react if the H atom it was approaching happened to vibrate and/or rotate towards it at the right time, thereby increasing the energy of collision. The F atom might be aimed at the centre of mass of the H2 molecule, but depending on the rotation of the molecule it might collide at the end or in the middle of the H2, or anywhere in between. The relatively static picture of a molecule sitting at the origin waiting for the on-centre approach of an attacking atom is a convenient simplification for the purpose of observing a reaction. If one wanted to perform such simulations in order to model a bulk reaction realistically, one would have to repeat the reaction many times with different initial conditions so as to obtain statistically meaningful results. (I did my M.Sc. on trajectory calculations of F + H2 --> HF + H. Sometimes it shows. :-) ) Regards, Joel ------------ Joel Polowin, Ph.D. Manager, Scientific Support Email to: polowin@hyper.com Hypercube Inc, 7-419 Phillip St, Waterloo, Ont, Canada N2L 3X2 (519)725-4040 Info requests to: info@hyper.com Support questions to: support@hyper.com Email group: Send "subscribe hyperchem" to hyperchem-request@hyper.com ________________________________________________________________________ From owner-hyperchem Thu Apr 6 12:21:42 1995 Date: Thu, 6 Apr 1995 08:23:02 -0400 X-Sender: underhil@137.94.5.141 To: hyperchem@hyper.com From: underhil@hp.rmc.ca (Ross Underhill) Subject: reactions There has been some interest in simulating reactions using semi-empirical methods. The best approach I've seen is that used in ab initio programs like Gaussian where you can do a geometry optimization looking for a single saddle point (i.e. a transition state). The program then allows the molecules to move down the slope on either side of the saddle point to trace the minimum energy trajectory. Trying to do this by hand (i.e. find the minimum energy configuration to get a reaction) would be very tedious for all but the simplest reactions (such as F + H2 = FH + H). One might ask Joel if Hyperchem has any intention of introducing this kind of capability. While I'm composing a note I would like to ask if the new ab initio capability in Hyperchem includes the calcluation of point charges by electrostatic potential fitting (i.e. routines like CHELPG). This would be a VERY useful feature since one of the requirements for Molecular Dynamics are ACCURATE atomic charges and I beleive it is generally agreed in the community that these can only be produced by good quality ab initio calculations. Hence the ability to do this would make the MD part of hyperchem much more complete. Dr. Ross Underhill Royal Military College of Canada Kingston, Ontario (613) 541-6000 X6175 ________________________________________________________________________ From owner-hyperchem Thu Apr 6 12:39:10 1995 Date: Thu, 6 Apr 95 14:20:24 +0100 From: "Ronald Wiegers" Subject: Error in CHEM.TPL: hydroxyproline To: hyperchem@hyper.com The template for 4-hydroxyproline (used in proteins like collagen) in Hyperchem release 4 for Windows is incorrect (error in template file). The following one works: ; 4-hydroxyproline [HYP]A N: N 1 (-3 s CA s CD s) CA: C 2 (N s HA s C s CB s) HA DA: H (CA s) \ opls none \ bio+ none C: C 3 (CA s O d +1 s) O: O 4 (C d) CB: C (CG s CA s 1HB s 2HB s) 1HB 1DB HB1 HB3: H (CB s) \ opls none \ bio+ none 2HB 2DB HB2: H (CB s) \ opls none \ bio+ none CG: C (CD s CB s OD1 s HG s) HG DG: H (CG s) \ opls none \ bio+ none OD1: O (CG s HOD s) HD1 DD1 HOD: H (OD1 s) CD: C (N s CG s 1HD2 s 2HD2 s) 1HD2 1DD2 HD21: H (CD s) \ opls none \ bio+ none 2HD2 2DD2 HD22: H (CD s) \ opls none \ bio+ none ===================================================== Regards, Ronald Wiegers Fuji Photo Film B.V. Tilburg, The Netherlands ronw@gpa.fuji-ef.nl ________________________________________________________________________ From owner-hyperchem Thu Apr 6 13:11:36 1995 Date: Thu, 6 Apr 1995 09:48:06 -0300 (BST) From: Edecio C Neto To: hyperchem@hyper.com Dear Hyperchemers, I am an M.D. with little background on hardcore Chemistry, just starting to use Hyperchem as a tool to predict shapes of antigenic oligopeptides. I would like to know about basic literature on molecular modeling/molecular dynamics of oligopeptides concerning both the parameters used (algorithm, time and temperature of run, dielectric constant, minimization in water, etc) that I can find in books or published papers. Gratefully, Edecio Cunha-Neto, M.D., Ph.D. Lab of Transplantation Immunology Faculty of Medicine, Universidade de Sao Paulo, Brazil ________________________________________________________________________ From owner-hyperchem Thu Apr 6 14:19:05 1995 Date: Thu, 06 Apr 1995 15:55:02 +0000 (GMT) From: KANEPCHEM <94970459@vax1.dcu.ie> Subject: Miscellaneous To: hyperchem@hyper.com Hi HyperChemers, I would be much obliged if anyone could answer the follwing questions about HyperChem: 1) I sometimes find that when carrying out an opmization on a molecule contain- ing a benzene ring that part of the ring is bent out of plane Firstly, is this sort of thing to be expected ? Secondly, I am interestedin the effect that making the ring planar again would have on the energy and any subsequent optimizations that I carry out. But how do I make the ring planar again without distorting the rest of the molecule ? The benzene rings which are being distorted are connected by -CH2- groups to form a larger ring. To make the benzene rings planar again, I have tried breaking one of the bonds to the -CH2- groups, highlighting the offending benzene ring and clicking on the Build/Add H and Model Build option. While this makes the ring planar again it also distorts the geometry in substituents attached to the benzene ring. 2) With regard to semi-empirical calculations a) what is meant by the following terms: binding energy, electronic energy and core-core interaction b) I carried out a AM1/single point calculation on benzene and obtained a value for isolated atomic energy of -18293.6 kcal/mol. I would have the isolated atoms would have 0 potential energy. So what is meant by the term isolated atomic energy c) are the values quoted for the heats of formation refer to standard conditions of 298 K and 100 kPa d) is it possible to calculate the Gibb's Free Energy of Formation and the Entropy of Formation e) Am I correct in saying that: Total Energy = Binding Energy + Isolated Atomic Energy = Electronic Energy + Core-Core Interaction Thanks in advance for any suggestions/answers Paddy Kane Dublin City University Ireland ________________________________________________________________________ From owner-hyperchem Thu Apr 6 19:28:53 1995 Date: Thu, 6 Apr 95 16:38:17 -0400 From: polowin@hyper.hyper.com (Joel Polowin) To: underhil@hp.rmc.ca (Ross Underhill) Subject: Re: reactions Cc: hyperchem@hyper.com > Date: Thu, 6 Apr 1995 08:23:02 -0400 > From: underhil@hp.rmc.ca (Ross Underhill) > > There has been some interest in simulating reactions using > semi-empirical methods. The best approach I've seen is that used in ab > initio programs like Gaussian where you can do a geometry optimization > looking for a single saddle point (i.e. a transition state). The program > then allows the molecules to move down the slope on either side of the > saddle point to trace the minimum energy trajectory. Trying to do this by > hand (i.e. find the minimum energy configuration to get a reaction) would be > very tedious for all but the simplest reactions (such as F + H2 = FH + H). > One might ask Joel if Hyperchem has any intention of introducing this kind > of capability. It's something that we've thought about. Trying to do this sort of thing is computationally tricky. It's fairly straightforward to find a saddle point for a simple reaction, as above; finding that saddle point for the n-dimensional hypersurface of multiple bond lengths and angles is harder. Essentially one would need a second-order optimizer that searched for saddle points instead of minima. I suspect that it's feasible but I am not sure if there is a GOOD general solution. Obviously if one reduces the number of degrees of freedom of the problem it becomes more tractable. (Will programs like Gaussian let one find saddle points in the general case of a polyatomic system?) It's on the wish list. > While I'm composing a note I would like to ask if the new ab initio > capability in Hyperchem includes the calcluation of point charges by > electrostatic potential fitting (i.e. routines like CHELPG). This would be > a VERY useful feature since one of the requirements for Molecular Dynamics > are ACCURATE atomic charges and I beleive it is generally agreed in the > community that these can only be produced by good quality ab initio > calculations. Hence the ability to do this would make the MD part of > hyperchem much more complete. The forthcoming version of HyperChem uses Mulliken charge distribution calculations, not electrostatic potential fitting. The latter was tried, but we couldn't get it working to our satisfaction. This, too, is on the wish list and we hope to have it for a future release. Regards, Joel ------------ Joel Polowin, Ph.D. Manager, Scientific Support Email to: polowin@hyper.com Hypercube Inc, 7-419 Phillip St, Waterloo, Ont, Canada N2L 3X2 (519)725-4040 Info requests to: info@hyper.com Support questions to: support@hyper.com Email group: Send "subscribe hyperchem" (or "unsubscribe hyperchem"!) to hyperchem-request@hyper.com ________________________________________________________________________ From owner-hyperchem Thu Apr 6 19:38:21 1995 Date: Thu, 6 Apr 95 16:13:20 -0400 From: polowin@hyper.hyper.com (Joel Polowin) To: "Ronald Wiegers" Subject: Re: Error in CHEM.TPL: hydroxyproline Cc: hyperchem@hyper.com > Date: Thu, 6 Apr 95 14:20:24 +0100 > From: "Ronald Wiegers" > > The template for 4-hydroxyproline (used in proteins like collagen) in > Hyperchem release 4 for Windows is incorrect (error in template file). > The following one works: > > ; 4-hydroxyproline > CG: C (CD s CB s OD1 s HG s) ^^^ Changed from original 'OG', a non-existant atom... > HD1 DD1 HOD: H (OD1 s) ^^^ Changed from original 'CG', the wrong atom. I don't know how those slipped through, but thanks for the correction; we'll change that immediately. Regards, Joel ------------ Joel Polowin, Ph.D. Manager, Scientific Support Email to: polowin@hyper.com Hypercube Inc, 7-419 Phillip St, Waterloo, Ont, Canada N2L 3X2 (519)725-4040 Info requests to: info@hyper.com Support questions to: support@hyper.com Email group: Send "subscribe hyperchem" (or "unsubscribe hyperchem"!) to hyperchem-request@hyper.com ________________________________________________________________________ From owner-hyperchem Fri Apr 7 08:17:10 1995 From: Frank Menzel Subject: normalization of coordinates To: hyperchem@hyper.com Date: Fri, 7 Apr 1995 11:13:44 +0200 (MET DST) Cc: cawb@asterix.rz.tu-clausthal.de (W. Brockner) Dear Hyperchemis! The molecule builder in HyperChem is a nice tool, but for several reasons the optimized geometry should be aligned to reference axes. Both alignment functions 'Align Molecules' and 'Align Viewer' do not transfer the molecule to new coordinates. I wish to transfer the centre of symmetry to coordinates 0,0,0 and other atoms to planes. Just with those 'normalized' coordinates, e.g. a reliable vibrational assignment of bigger molecules with respect to 'ir-intensity-components' and 'ir-normal-mode' variables (from script) is possible. Is there any possibility to rotate/normalize the molecule (and possibly the displacement vectors too)? Thanks Frank Menzel ________________________________________________________________________ From owner-hyperchem Mon Apr 10 12:37:42 1995 From: "Jens Spanget-Larsen" Organization: Roskilde Universitetscenter To: polowin@hyper.hyper.com (Joel Polowin) Date: Mon, 10 Apr 1995 13:17:01 +0100 Subject: Re: reactions Cc: hyperchem@hyper.com > > Date: Thu, 6 Apr 1995 08:23:02 -0400 > > From: underhil@hp.rmc.ca (Ross Underhill) > > > > There has been some interest in simulating reactions using > > semi-empirical methods. The best approach I've seen is that used in ab > > initio programs like Gaussian where you can do a geometry optimization > > looking for a single saddle point (i.e. a transition state). The program > > then allows the molecules to move down the slope on either side of the > > saddle point to trace the minimum energy trajectory. Trying to do this by > > hand (i.e. find the minimum energy configuration to get a reaction) would be > > very tedious for all but the simplest reactions (such as F + H2 = FH + H). > > One might ask Joel if Hyperchem has any intention of introducing this kind > > of capability. > > It's something that we've thought about. Trying to do this sort of thing > is computationally tricky. It's fairly straightforward to find a saddle > point for a simple reaction, as above; finding that saddle point for the > n-dimensional hypersurface of multiple bond lengths and angles is harder. > Essentially one would need a second-order optimizer that searched for > saddle points instead of minima. I suspect that it's feasible but I am > not sure if there is a GOOD general solution. Obviously if one reduces > the number of degrees of freedom of the problem it becomes more tractable. > (Will programs like Gaussian let one find saddle points in the general case > of a polyatomic system?) It's on the wish list. > Regarding saddle points: I just want to mention that in MOPAC, "general" procedures are implemented that allow determination of saddle points corresponding to transition states (keywords SADDLE and TS). For example, to use the SADDLE option, you input two geometries, one corresponding to the reactants and one to the products, and MOPAC then searches for a transition state. The resulting transition state structure can be refined by using the TS option. I have used these MOPAC procedures for several polyatomic systems with great success; but, of course, the SADDLE procedure occasionally breaks down for complicated reactions with multiple reaction paths, etc. Yours, Jens >--< =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= JENS SPANGET-LARSEN Department of Chemistry Phone: +45 46757711 Roskilde University (RUC) Fax: +45 46757721 POB 260, DK-4000 Roskilde, Denmark Internet: JSL@mmf.ruc.dk =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= ________________________________________________________________________ From owner-hyperchem Tue Apr 11 07:19:59 1995 From: Piotr Skurski Subject: Saddle points To: hyperchem@hyper.com Date: Tue, 11 Apr 95 8:31:30 METDST Cc: JSL@mmf.ruc.dk > > Date: Thu, 6 Apr 1995 08:23:02 -0400 > > From: underhil@hp.rmc.ca (Ross Underhill) > > > > There has been some interest in simulating reactions using > > semi-empirical methods. The best approach I've seen is that used in ab > > initio programs like Gaussian where you can do a geometry optimization > > looking for a single saddle point (i.e. a transition state). The program > > then allows the molecules to move down the slope on either side of the > > saddle point to trace the minimum energy trajectory. Trying to do this by > > hand (i.e. find the minimum energy configuration to get a reaction) would be > > very tedious for all but the simplest reactions (such as F + H2 = FH + H). > > One might ask Joel if Hyperchem has any intention of introducing this kind > > of capability. > > It's something that we've thought about. Trying to do this sort of thing > is computationally tricky. It's fairly straightforward to find a saddle > point for a simple reaction, as above; finding that saddle point for the > n-dimensional hypersurface of multiple bond lengths and angles is harder. > Essentially one would need a second-order optimizer that searched for > saddle points instead of minima. I suspect that it's feasible but I am > not sure if there is a GOOD general solution. Obviously if one reduces > the number of degrees of freedom of the problem it becomes more tractable. > (Will programs like Gaussian let one find saddle points in the general case > of a polyatomic system?) It's on the wish list. > Regarding saddle points: I just want to mention that in MOPAC, "general" procedures are implemented that allow determination of saddle points corresponding to transition states (keywords SADDLE and TS). For example, to use the SADDLE option, you input two geometries, one corresponding to the reactants and one to the products, and MOPAC then searches for a transition state. The resulting transition state structure can be refined by using the TS option. I have used these MOPAC procedures for several polyatomic systems with great success; but, of course, the SADDLE procedure occasionally breaks down for complicated reactions with multiple reaction paths, etc. Yours, Jens >--< =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= JENS SPANGET-LARSEN Department of Chemistry Phone: +45 46757711 Roskilde University (RUC) Fax: +45 46757721 POB 260, DK-4000 Roskilde, Denmark Internet: JSL@mmf.ruc.dk =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Dear HyperChem'ers, About saddle points - there is a powerful program called GAMESS (General Atomic and Molecular Electronic Structure System) which also allows locating a transition state (semiempirical and ab initio level). I have used this program many times. It works ! Important: Copies of GAMESS are provided at no charge ! Yours, Piotr Piotr Skurski, M.Sc. Department of Chemistry University of Gdansk Sobieskiego 18, Gdansk POLAND e-mail: piotr@ewa.chem.univ.gda.pl ________________________________________________________________________ From wtwinter@mailbox.syr.edu Mon Apr 10 01:14:06 1995 Date: Mon, 10 Apr 1995 01:20:39 -0400 (EDT) From: "William T. Winter" X-Sender: wtwinter@mothra.syr.edu To: compchem Cc: Hyperchem Subject: Polysaccharide Coordinate Database I have begun to see several postings for coordinates of diffraction based polysaccharide structures. A few such entries do exist in the pdb ( Protein Data Base). These include chondroitin sulfates, iota carrageenan, some hyaluronates, and a capsular polysaccharide from a strain of E. coli. All of these were submitted by the Arnott group during the late 1970's. I was a member of that group then and continue to be active in such studies as does my colleague and department head Anatole Sarko. If you would like to see a separate polysaccharide date base developed, send me a short note of support by email, FAX or regular mail. If you have preferences as to form - pdb format or extension thereof, Cartesian or fractional unit cell etc - let me know about those preferences I am coming into a period where I will have the time to seek support for and initiate development of such a data base but I am certain that no support will be forthcoming unless I can demonstrate a potential community of users. Finally, if such a data base were extended to include other linear or regularly branched polymers both naturally occurring and synthetic would that heighten the interest? I think that the nucleic acid people may already have such a data base but I do not believe that the synthetic polymer people have gotten to this stage yet. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Dr. William T. Winter Phone: (315)470-6876 315 Baker Lab FAX: (315)470-6856 SUNY-ESF Internet: wtwinter@mailbox.syr.edu Syracuse, NY 13210-2786 ________________________________________________________________________ From owner-hyperchem Tue Apr 11 15:26:41 1995 From: curtisj@usfca.edu Date: Mon Apr 10 07:47:30 1995 To: hyperchem@hyper.com Subject: display of Gaussian results Dear Hyperchem World, Has anyone out there found a way to display the orbital results from an ab-initio calculation done in Gaussian-for-Windows from within Hyperchem or Chemplus? Much obliged, -- Jeff Curtis at University of San Francisco ________________________________________________________________________ From owner-hyperchem Tue Apr 11 18:54:24 1995 Date: Tue, 11 Apr 95 16:16:19 -0400 From: polowin@hyper.hyper.com (Joel Polowin) To: Frank Menzel Subject: Re: normalization of coordinates Cc: hyperchem@hyper.hyper.com > From: Frank Menzel > Date: Fri, 7 Apr 1995 11:13:44 +0200 (MET DST) > > The molecule builder in HyperChem is a nice tool, but for several reasons > the optimized geometry should be aligned to reference axes. Both alignment > functions 'Align Molecules' and 'Align Viewer' do not transfer the molecule > to new coordinates. I wish to transfer the centre of symmetry to coordinates > 0,0,0 and other atoms to planes. Just with those 'normalized' coordinates, > e.g. a reliable vibrational assignment of bigger molecules with respect to > 'ir-intensity-components' and 'ir-normal-mode' variables (from script) is > possible. > Is there any possibility to rotate/normalize the molecule (and possibly > the displacement vectors too)? If you select a molecule, you can then use Edit/Translate to move that selection to the origin; the centre of mass of the selection ends up at the origin. I am not sure if there is a relatively easy way to rotate the system so as to bring another atom into a specific plane. Given the x,y,z coordinates of an atom (by selecting it, or from a script) it is a matter of fairly simple trigonometry to determine what rotation is necessary to bring that atom into the xy, yz, or xz plane. A macro in Excel or a Visual Basic program to do such a calculation and rotation should be fairly straightforward. Regards, Joel ------------ Joel Polowin, Ph.D. Manager, Scientific Support Email to: polowin@hyper.com Hypercube Inc, 7-419 Phillip St, Waterloo, Ont, Canada N2L 3X2 (519)725-4040 Info requests to: info@hyper.com Support questions to: support@hyper.com Email group: Send "subscribe hyperchem" to hyperchem-request@hyper.com ________________________________________________________________________ From owner-hyperchem Wed Apr 12 08:38:32 1995 From: "Jens Spanget-Larsen" Organization: Roskilde Universitetscenter To: hyperchem@hyper.com Date: Wed, 12 Apr 1995 11:31:01 +0100 Subject: Re: normalization of coordinates > Dear Hyperchemis! > > The molecule builder in HyperChem is a nice tool, but for several reasons > the optimized geometry should be aligned to reference axes. Both alignment > functions 'Align Molecules' and 'Align Viewer' do not transfer the molecule > to new coordinates. I wish to transfer the centre of symmetry to coordinates > 0,0,0 and other atoms to planes. Just with those 'normalized' coordinates, > e.g. a reliable vibrational assignment of bigger molecules with respect to > 'ir-intensity-components' and 'ir-normal-mode' variables (from script) is > possible. > Is there any possibility to rotate/normalize the molecule (and possibly > the displacement vectors too)? > > Thanks > > Frank Menzel > Just one little "trick" that I have found useful: Save the structure as a Z-matrix, *.ZMT. When the structure is opened from this file format, atom no. 1 is placed at the origin (0,0,0) of the cartesian coordinate system, atom no. 2 is on the x-axis (X2,0,0), and atom no. 3 in the x,y-plane (X3,Y3,0). Jens >--< =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= JENS SPANGET-LARSEN Department of Chemistry Phone: +45 46757711 Roskilde University (RUC) Fax: +45 46757721 POB 260, DK-4000 Roskilde, Denmark Internet: JSL@mmf.ruc.dk =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= ________________________________________________________________________ From owner-hyperchem Thu Apr 13 18:02:39 1995 From: "Dr. Steven Desjardins" Organization: Washington & Lee University To: hyperchem@hyper.com Date: Thu, 13 Apr 1995 15:02:58 EDT Subject: Royalties for manuals Hi. I am teaching a course which has a section on molecular modelling, and I think that sections of the HyperChem computational chemistry manual would make a good introduction. As I don't want to have students all buy a manual set, could I make copies of certain sections and pay HyperCube a royalty? The course has about 25 students. ________________________________________________________________________ From owner-hyperchem Wed Apr 19 16:21:54 1995 Date: Wed, 19 Apr 1995 18:29:46 +0000 (GMT) From: KANEPCHEM <94970459@vax1.dcu.ie> Subject: Previous Message To: hyperchem@hyper.com Hi Joel, A few weeks ago, I posted a message to the user group, mainly about semi- empirical calculations but as yet, I have received no reply. Can you tell me if you noticed such a message ? Paddy Kane ________________________________________________________________________ From owner-hyperchem Thu Apr 20 17:04:42 1995 Date: Thu, 20 Apr 95 13:34:55 -0400 From: polowin@hyper.hyper.com (Joel Polowin) To: KANEPCHEM <94970459@vax1.dcu.ie> Subject: Re: Previous Message Cc: hyperchem@hyper.hyper.com > Date: Wed, 19 Apr 1995 18:29:46 +0000 (GMT) > From: KANEPCHEM <94970459@vax1.dcu.ie> > Subject: Previous Message Hi, Paddy. > A few weeks ago, I posted a message to the user group, mainly about semi- > empirical calculations but as yet, I have received no reply. Can you tell > me if you noticed such a message ? Yes; it got buried in my stack. I was hoping that someone else who's more familiar with semi-empirical calculations than I am might answer your general questions, or at least point you (and others on the list) towards a couple of good general reference texts on the subject. (The _Computational Chemistry_ manual suggests Levine's _Quantum Chemistry_ and Szabo and Ostlund's _Modern Quantum Chemistry_.) As for your specific questions: > 1) I sometimes find that when carrying out an opmization on a molecule > containing a benzene ring that part of the ring is bent out of plane > Firstly, is this sort of thing to be expected ? This depends on your system. A free benzene ring should remain planar. If something is coordinated to it, then the resulting interactions could easily cause the six attachments to bend out of the C6 plane, either towards or away from the coordinating species; if the coordination is not centered on the C6 hexagon, or if the coordinating group is bulky, then the distortions may not be symmetrical and even the C6 may not remain planar. Finally, if the six attachments are bulky or otherwise in an awkward conformation, they might cause the benzene to be non-planar. You don't specify how non-planar your resulting structures are, but you've got to remember that in any sort of molecular modelling, the final structures result from compromises between many interactions to give an energy minimum. There are forces that are trying to make the aromatic ring planar; there may be other forces that tend to cause deviations from planarity. *ANY* non-planar forces *will* cause the benzene to be non- planar; the question is whether that deviation is tiny or noticeable. The minimum-energy conformation according to the ring forces alone is planar, and the force constants for those interactions are relatively high. But even a tiny out-of-plane force will distort the system a little bit, until the forces balance in a combined energy minimum. > Secondly, I am interestedin the effect that making the ring planar again > would have on the energy and any subsequent optimizations that I carry out. > But how do I make the ring planar again without distorting the rest of the > molecule ? The benzene rings which are being distorted are connected by > -CH2- groups to form a larger ring. To make the benzene rings planar again, > I have tried breaking one of the bonds to the -CH2- groups, highlighting the > the offending benzene ring and clicking on the Build/Add H and Model Build > option. While this makes the ring planar again it also distorts the geometry > in substituents attached to the benzene ring. The model builder is intended to give a "quick-and-dirty" three-dimensional structure, which should then be refined with one of the modelling techniques. Running it on part of a structure after an optimization will cause that kind of distortion. Forcing part of a structure into a particular geometry -- planarity, for example -- against opposing forces must necessarily result in distortion somewhere. In this case, the distortions may be in the C6 attachments themselves, or in the angles with respect to their connections to the C6. If you want to enhance the planarity of your C6 rings, you can bias the balance of forces towards that configuration. You could modify the torsional force constants, increasing the 2-periodic values for aromatic atoms (i.e., the forces which favour planarity). But it would probably be simpler to add a restraint to each of the six torsional angles around the C6 ring, specifying torsional angles of 0', to boost the forces that are tending to planarity. That way, after structural optimization, you could remove the restraints and calculate the single-point energy; the difference between that and the original optimized structure's energy would give you the energy effects of forcing planarity on the system. By repeating this several times, using different force constants for the restraints, you could investigate the specific energy effects of different degrees of planarity. If you wanted the attachments also to be in the C6 plane, you could use other restraints involving their atoms and the C6 atoms. > 2) With regard to semi-empirical calculations > a) what is meant by the following terms: > binding energy, electronic energy and core-core interaction > b) I carried out a AM1/single point calculation on benzene and obtained a > value for isolated atomic energy of -18293.6 kcal/mol. I would have > the isolated atoms would have 0 potential energy. So what is meant by the > term isolated atomic energy > c) are the values quoted for the heats of formation refer to standard > conditions of 298 K and 100 kPa > d) is it possible to calculate the Gibb's Free Energy of Formation > and the Entropy of Formation > e) Am I correct in saying that: > Total Energy = Binding Energy + Isolated Atomic Energy > = Electronic Energy + Core-Core Interaction I'll leave most of this for others, since my background in the fundamentals of semi-empirical calculations is quite limited. The quoted values for heats of formation do refer to standard conditions. Computing internal entropies for these systems is generally not feasible since, for example, the rotational entropy of groups within a structure depends critically on interactions with non-bonded neighbors that vary enormously with bending, stretching, etc. Essentially, it fundamentally involves dynamic factors, where the modelling process finds static minima. Regards, Joel ------------ Joel Polowin, Ph.D. Manager, Scientific Support Email to: polowin@hyper.com WWW: http://www.hyper.com/ Hypercube Inc, 7-419 Phillip St, Waterloo, Ont, Canada N2L 3X2 (519)725-4040 Info requests to: info@hyper.com Support questions to: support@hyper.com Email group: Send "subscribe hyperchem" (or "unsubscribe hyperchem") to hyperchem-request@hyper.com; please do not send such administrative messages to the group itself. ________________________________________________________________________ From owner-hyperchem Mon Apr 24 19:50:30 1995 Date: Mon, 24 Apr 95 15:59:22 EDT From: "Franklin V. Cobos II" Subject: Co Complexes - UV spectra To: hyperchem@HYPER.COM I'm new to hyperchem, though not to molecular modeling, and am trying to do some UV spectra predictions with it...since the other packages that I've used in the past didn't have that feature. Does anyone have any experience with predicting spectra of coordination complexes? Please respond direct: fcobos@utcvm.utc.edu ________________________________________________________________________ From owner-hyperchem Tue Apr 25 15:17:52 1995 Date: Tue, 25 Apr 1995 17:48:52 +0000 (GMT) From: KANEPCHEM <94970459@vax1.dcu.ie> Subject: Miscellaneous To: hyperchem@hyper.com Dear HyperChemers Can anyone tell me if it possible to calculate entropy of formation and therefore Gibb's Free Energy of Formation ? Thanks in advance for any suggestions Regards, Paddy Kane Dublin City University Ireland ________________________________________________________________________ From owner-hyperchem Wed Apr 26 09:17:15 1995 Date: Wed, 26 Apr 1995 11:52:57 UTC+0100 From: "Jose G. de la Campa" Subject: About windows 95 To: hyperchem@www.hyper.com We have just installed the final beta release of windows 95. Everything seems to work properly in Hyperchem except for ChemPlus. If we draw a molecule in Hyperchem and select presentatons to go to ChemPlus, we get a message saying "Query-value: Unknown variable 'atomic number' and after that we get all the atoms in red color, what means that ChemPlus does not recognize the type of atom. We have checked that all the settings are the same that we had previously in Windows 3.11. Therefore, we suppose it is a problem of DDE. Does windows 95 support DDE or only OLE? Is ther any way to solve this problem? Jose G. de la Campa, Instituto de Polimeros, Madrid ________________________________________________________________________ From owner-hyperchem Wed Apr 26 16:20:13 1995 X-Sender: jfroehli@pop.tuwien.ac.at To: hyperchem@hyper.com From: jfroehli@pop.tuwien.ac.at (Dr. Johannes Froehlich) Subject: ChemPlus Problems Cc: support@hyper.com Date: Wed, 26 Apr 1995 17:00:22 +0000 >To: polowin@hyper.com (Joel Polowin) >From: jfroehli@pop.tuwien.ac.at (Dr. Johannes Froehlich) >Subject: ChemPlus Problems >Cc: >Bcc: >X-Attachments: > >Jose G. de la Campa wrote: >>We have just installed the final beta release of windows 95. Everything seems to work properly in hyperchem except for ChemPlus. If we draw a molecule in Hyperchem and select presentatons to go to chemPlus, we get a message saying "Query-value: Unknown var >>iable 'atomic number' and after that we get all the atoms in red color, what means that chemplus does not recognize the type of atom. We have checked that all the settings are the same that we had previously in Windows 3.11. Therefore, we suppose it is a p >>roblem of DDE. Does windows 95 support DDE or only OLE? Is ther any way to solve this problem? >>Jose G. de la Campa, Instituto de Polimeros, Madrid >> >> > >We experienced a similar problem with Chemplus running under Windows NT 3.5 when using the Rainbow-NT-Dongle Driver (freely obtained by email from HyperCube). > >Will this problem be fixed by HyperCube? > >Regards: >Johannes Froehlich > >*************** >Dr. Johannes Froehlich >Techn. Univ. Vienna, Inst. Org. Chem. >email: jfroehli@pop.tuwien.ac.at >*************** > > ________________________________________________________________________ From owner-hyperchem Thu Apr 27 15:27:26 1995 Date: Thu, 27 Apr 95 12:13:07 -0400 From: polowin@hyper.hyper.com (Joel Polowin) To: Macromol-2@PINAR1.csic.es, tao@ecn.purdue.edu, jfroehli@pop.tuwien.ac.at Subject: Re: ChemPlus Problems, Windows 95 Cc: hyperchem@www.hyper.com Jose G. de la Campa wrote: > We have just installed the final beta release of windows 95. Everything > seems to work properly in hyperchem except for ChemPlus. If we draw a > molecule in Hyperchem and select presentatons to go to chemPlus, we get a > message saying "Query-value: Unknown variable > 'atomic number' and after that we get all the atoms in red color, > what means that chemplus does not recognize the type of atom. We have > checked that all the settings are the same that we had previously in Windows > 3.11. Therefore, we suppose it is a problem of DDE. Does windows 95 > support DDE or only OLE? Is ther any way to solve this problem? Windows 95 supports both. We *thought* that operations via DDE under Windows 95 would be the same as before, though we hadn't actually tested the HyperChem--ChemPlus interaction. The problems that you describe are characteristic of DDE problems, yes -- the same sort of thing that we've seen under Windows when the International settings make the DDE do strange things to data going through it. (I assume that you have the decimal and thousands separators as the period and comma, respectively, and that if you are working in Spanish you are using "Spanish (Modern)" and not just "Spanish" as the language setting.) Johannes Froehlich wrote: > We experienced a similar problem with Chemplus running under Windows NT 3.5 > when using the Rainbow-NT-Dongle Driver (freely obtained by email from > HyperCube). > > Will this problem be fixed by HyperCube? Bernie Tao wrote: > Ditto from our labs on problems with NT. > The hyperchem folks at the ACS mtg in Anaheim said that they did not have any > plans currently to have an NT version availbale. ChemPlus doesn't work at all under NT; it *should*, but it doesn't, and we don't know why yet. We don't have immediate plans for an NT version, nor for a native NT version of HyperChem since the Windows version runs correctly (apart from needing the special port driver). If we can't get ChemPlus for Windows to run under NT, we may have to reconsider this. (We *do* have an NT version of HyperChem for the DEC Alpha.) Question to the group: Has anyone got HyperChem to interact with ChemPlus or HyperNMR correctly under Windows 95 or under Windows NT? How about running HyperChem via external software such as Visual Basic or an Excel macro, under these systems? I'm hoping that solving these problems might be a matter of making sure that some settings in the operating environment are correct (i.e., compatible with our software), rather than having to change the software. Release 5 of HyperChem *will* be fully compatible with Windows 95. The official press release about Release 4.5 notes that we didn't want to hold up making _ab initio_ calculations available to our users until after Windows 95 ships, so we're putting out the intermediate 4.5 version. (A couple of people have asked me when Windows 95 would be out, and all I can say is that we're sort of expecting that it will be sometime in '95.) Regards, Joel ------------ Joel Polowin, Ph.D. Manager, Scientific Support Email to: polowin@hyper.com WWW: http://www.hyper.com/ Hypercube Inc, 419 Phillip St, Waterloo, Ont, Canada N2L 3X2 (519)725-4040 Info requests to: info@hyper.com Support questions to: support@hyper.com Email group: Send "subscribe hyperchem" (or "unsubscribe hyperchem") to hyperchem-request@hyper.com; please do not send such administrative messages to the group itself. ________________________________________________________________________ From owner-hyperchem@hyper.com Fri Apr 28 09:29:21 1995 From: David Simpson Subject: peptide conformation To: hyperchem@hyper.com (hyperchem) Date: Fri, 28 Apr 1995 09:18:02 +0200 (MET DST) Some weeks ago there was an enquiry about using Hyperchem to calculate the conformation of synthetic oligopeptides used as antigens. I did not see any replies, and I would like to hear from anyone who has had experience in this field. It seems to me that the solution depends on the initial guess, and after energy minimisation, there is very little difference (in kcals) between the different structures. David Simpson Dept. of Physiology, Carlsberg Lab Copenhagen ________________________________________________________________________ From owner-hyperchem@hyper.com Fri Apr 28 19:36:52 1995 From: dsmith@doane.edu Date: Fri, 28 Apr 1995 14:16:50 -0500 To: hyperchem@hyper.com Subject: wierd printing Hyperchemers! As an introduction to modeling and to get my students more used to seeing molecules in 3-D, I have them draw the structures of the compounds they use in lab on Hyperchem, play with them a bit, then print them for their lab report as framework drawings and in spheres. One student discovered a new feature of Hyperchem (or of our computer system, and we can't find how to *un*discover it: When we draw a molecule, add H and model build, and render it in spheres, it will print ok the first time. Thereafter, the molecule is split into 3 parts. For instance, on the second printing the molecule should be 70 mm in length, but the top 6 mm are cut off and printed on the *bottom*, and the bottom 6 mm are printed on the top. The two fragments are printed about 13 mm from the bulk of the molecule. Subsequent attempts to print get worse--on the fourth printing the molecule is divided into thirds. If we clear out this molecule and draw a new one, the cycle begins again--a good first printing... It didn't useta do this before! Any ideas for making Hyperchem (or the computer) not do this again? Thanks! Dave Smith Doane College Crete NE 68333 dsmith@doane.edu