From chemistry-request@ccl.net Mon Mar 18 11:36:18 1991 Date: Mon, 18 Mar 91 16:12:39 GMT From: jpj@lotus.medicine.rochester.EDU Subject: Partition Coefficient Calculations To: CHEMISTRY@ccl.net Status: R Recently, commercial sources have become available for programs that use semiempirical results to compute octanol/water partition coefficients. Is this type of software available in the public domain? Does it work? Has anybody compared packages from different vendors? jpj@lotus.medicine.rochester.edu From chemistry-request@ccl.net Mon Mar 18 17:44:57 1991 Date: Mon, 18 Mar 91 14:11:59 EST From: Shaun Black Subject: Partition Coefficient Calculators To: "Jeffrey P. Jones" Status: R I have some information that may be useful. First, there was an article in C&E News (9/19/88, volume 27) that described a new firm CompuDrug USA of Austin, Texas. They were marketing a program "Pro-logp" which is an expert system approach to estimation of the hydrophobicity of a compound from its structure. It runs on IBM/compatibles and VAX systems. Second, we have just published a new set of log(P) (o/w) based params for the standard twenty amino acids and for about twenty post- or co-translationally modified derivatives (Black, S.D. and Mould, D.R. (1991) Anal. Biochem. 193: 72-82). Our approach correlates very well with empirical studies of others (for the standard twenty) and would seem to be a reasonable way into various aspects of protein polarity/hydrophobicity. I hope this will be helpful to you. Enjoy! =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Shaun D. Black | Bitnet: black@ohstphrm.bitnet Ohio State University | Internet: black@ohstphrm.pharmacy.ohio-state.edu College of Pharmacy | Phone: (614) 292-3925 500 West 12th Avenue | FAX: (614) 292-2435 Columbus, OH 43210-1291 | :-) =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From chemistry-request@ccl.net Mon Mar 18 18:47:56 1991 Date: Mon, 18 Mar 91 17:28 CDT From: Subject: MOPAC vs AMPAC To: chemistry@ccl.net Status: R Someone was wondering why MOPAC is so much more popular than AMPAC.... In my case, I use MOPAC for anything that starts far from a minimum: back in the days of MOPAC v4 I found that MOPAC had less tendency to "crash" with less-than-optimally-defined inputs. In cases where both crashed, I found that MOAPC gave me more information that I could use to redefine the problem & get to a result I wanted. However, because it runs faster, I use AMPAC for things like computing the force constants of an already-optimized structure. I recently ran a comparison of a force computation with mopac v6, and what the Cray I'm running on refers to as the '90 edition of ampac: mopac took 8xx cpu seconds, while ampac took 5xx sec. It may well be that this difference between mopac and ampac has been programmed out, but I don't have enough cpu time lying around to want to do the experiment all over again. Irene Newhouse