From chemistry-request@ccl.net Tue Mar 31 00:01:18 1992 Date: Mon, 30 Mar 92 19:18:19 PST From: raman@scripps.EDU (K. Ramnarayan) Subject: Antibody Modelling To: chemistry@ccl.net Status: R Hello comp. chemists, I am looking for a suitable program (both public domain or commercial) to perform antibody modelling, specifically the modelling of the CDRs. You can respond to me directly at raman@scripps.edu. I will summarize the responses to the net. Thank you in advance. From chemistry-request@ccl.net Tue Mar 31 14:06:18 1992 Date: Tue, 31 Mar 92 09:55:40 PST From: whitbeck@wheeler.wrc.unr.EDU (Mike Whitbeck) Subject: rovibrational band width To: chemistry@ccl.net Status: R Dear Computational Chemists: I am interested in relating the bandwidth of -OH strecting (ca 3500-3700 cm-1) of a solid polyol to rotation about the carbon-oxygen axis, i.e. rotation of -OH's through H-bonded and non-H-bonded positions: H / axis A heat / C-----------O =========>> -------O \ \ H .......O--- o---- =========1================= ===========2=========== both (1) and (2) are crystals in a KBr matrix. Starting with (1) at low temperature the OH str corresponds to an H-BONDED configuration. On heating there is little change up to the point of phase transition then there is a sharp frequency shift corresponding to the non-H-Bonded configuration (2). Thereafter the the band continues to grow in width with increasing temperature. It also continues to shift to higher frequency although only weakly so. I believe the increase in width may be explained in terms of the angular velocity of the H as it rotates about its' C-O axis. Given that this is a macroscopic crystal lattice as opposed to an isolated molecule is it feasible to calculate a band width using a 'local group' approximation for this internal rotation? Any references come to mind? Would such a calculation be a major undertaking or trivial? [I generally measure spectra rather than calculate them :-) but this is a case where modeling would be of considerable benefit] I'd appreciate any comments or tips on the above. Thanks in advance. Mike Whitbeck Desert Research Inst. whitbeck@wrc.unr.edu All academics have the potential for being insatiable... but the chemists are the most expensive and insatiable among the expensive and insatiable. - J. Martin in "To Rise Above Principle" From chemistry-request@ccl.net Tue Mar 31 15:01:19 1992 Date: Tue, 31 Mar 92 12:59:50 EST From: mitchell@bdrc.bd.COM (Mike Mitchell) Subject: Alliant FX/40 for sale or ? To: chemistry@ccl.net Status: R We have an Alliant FX/40 with 3 processors and 6 Fujitsu drives (3Gb total) that we are phasing out. We would like to sell this machine or perhaps donate it to a worthy academic institution. Any offers? Mike Mitchell mitchell@bdrc.bd.com From chemistry-request@ccl.net Tue Mar 31 16:12:35 1992 Date: Tue, 31 Mar 92 13:29:33 CST From: schneid@csrd.uiuc.EDU (David John Schneider) Subject: rovibrational band width To: whitbeck@wheeler.wrc.unr.EDU Status: R > Return-Path: > Date: Tue, 31 Mar 92 09:55:40 PST > From: whitbeck@wheeler.wrc.unr.edu (Mike Whitbeck) > To: chemistry@ccl.net > Subject: rovibrational band width > Sender: chemistry-request@ccl.net > Precedence: bulk > > Dear Computational Chemists: > > I am interested in relating the bandwidth of -OH strecting (ca 3500-3700 cm-1) > of a solid polyol to rotation about the carbon-oxygen axis, i.e. > rotation of -OH's through H-bonded and non-H-bonded positions: > > [text deleted] > Mike, If I am not mistaken, Roy Gordon and others developed the theory of infrared and Raman lineshapes back in the 60's and early 70's, much along the same lines as was previously done for magnetic resonance lineshape theory. Basically, what the theory requires is the calculation of various time correlation functions for the dipole moment. The Fourier transform of the correlation function gives the lineshape in the frequency domain. Though this sort of modelling is most commonly done for liquids, it can also be applied to solids. Similar problems have been studied by magnetic resonance, where "local group" approximations to the autocorrelation functions is often employed. Dave Schneider University of Illinois at Urbana-Champaign Center for Supercomputing Research and Development 317 Talbot Laboratory 104 S. Wright Street Urbana, IL 61801 phone : (217) 244-0055 fax : (217) 244-1351 E-mail: schneid@csrd.uiuc.edu From chemistry-request@ccl.net Tue Mar 31 23:51:42 1992 Date: Tue, 31 Mar 92 21:25 CDT From: NEELY%AUDUCVAX.BITNET@OHSTVMA.ACS.OHIO-STATE.EDU Subject: user-defined residues in AMBER 3.0 To: chemistry@ccl.net Status: R I'm trying to model molecular interactions between phospholipids and other molecules using AMBER 3.0. However, I've been stumped trying to get my user-defined residues to link into a phospholi- pid properly. Starting from infinite ignorance vis-a-vis AMBER, I started by splitting my molecule into 3 parts: the choline headgroup, the phosphate, and the rest. I am using a PDB-style cartesian coordinate file that I generated from a combination of model-building utilities. It plots as a reasonable molecule. Each of the residues, for which I also used cartesian coordinates, also plots as a chemically reasonable molecule. By dint of much trial & error, I managed to get the choline & phosphate to LINK when one of the dummy atoms in each residue is at the position the next atom is to bind. The problem is, that the third residue LINKS at this same position! I have consulted Bill Ross with regard to this problem, and he has suggested I query the mailer to see if anyone else has come across this problem. Is there anyone out there who can help me? How have other people defined lipid molecules for AMBER? Thanks for any help! Irene Newhouse