From chemistry-request@ccl.net Wed May 20 09:12:58 1992 Date: Tue, 7 Apr 92 09:56:12 BST From: Mark Forster To: chemistry Status: R Subject: Movies Dear Comp. Chemists. Following the recent discussions on movie file formats, I would like to ask people about their experiences of transferring movies to video tape. The questions to be posed are 1. What workstation or PC system holds the coordinate/connectivity data. 2. What hardware is required to interface this to the video cassette recorder. 3. How does this differ for the different TV picture formats, i.e. NTSC in the USA, PAL in the UK, SECAM in Europe. 4. How much does 2. cost and what is the quality of the results. I would welcome commercial plugs answering these questions because these are the systems we will probably end up purchasing. For the benefit of chemically oriented users please assume a minimum level of hardware knowledge. Best Wishes Mark.Forster forster@comp.nibsc.ac.uk ps. sending message to CHEMISTRY%ccl.net@uk.ac.nsfnet-relay From chemistry-request@ccl.net Wed May 20 23:23:34 1992 Date: Wed, 20 May 92 18:30:33 PDT From: whitbeck@wheeler.wrc.unr.EDU (Mike Whitbeck) Subject: IR spectral interpretation...announcement To: chemistry@ccl.net Status: R If you teach chemistry and/or infrared spectroscopy the following may be of interest. I have just written a program that assists in the interpretation of infrared spectra (see note 0 below). The program is in the testing phase now and when that is complete I will make it available by anonymous ftp with an anouncement here. As with KINETICS2 I will provide source code and permit the free use and distribution of the code for academic and other non-commercial use. I do this so that the code itself can be used instructionally and so you can easily modify/customize to your heart's desire. I WOULD LIKE to identify some suitable 'beta-testers' (read guinea pigs ;-) with knowledge of infrared spectroscopy and C++, access to a suitable machine (e.g. sun sparc... probably too big for a pc, access to ftp, access to g++ or suitable c++ compiler) etc. IF INTERESTED send me some email (whitbeck@wrc.unr.edu or whitbeck@equinox.unr.edu). BUT... be forewarned the code is not small! Here are some pertinent facts: 1. coded in C++ 2. compiled/tested only with GNU g++ and only on a SUN sparc IPC (a Unix box) 3. may be difficult to port to IBM PC's (but not impossible) 4. uses the LEDA class library [ see note 2 below ], available by ftp features: 1. reads and uses both peak position and intensity data(absorbance) 2. reads user selected group frequency table (170 provided) 3. provides simple editing of group frequency tables to filter out unwanted entries 4. algorithm based... not a bunch of IF () THEN frames 5. output includes information on how the code reached the assignments that were made! 6. fuzzy membership is used in relating peaks to assignments 7. very fast (at least on my sparc, see note 3) An example is given below (see note 3). ------------------------------ note 0: By 'assist' I mean that the program is meant to be used as a tool/engine; while it can be used, in a limited fashion as-is for spectral interpretation (see the example below) my intent is to provide code that can be modified and extended to produce 'customized' spectral interpretation. Hopefully this customization will lead to better algorithms (and their publication). At present this customization can be performed to a limited extent external to the program without a programming knowledge however the best and preferred method would be through additional C++ code utilizing the basic engine- hence source code disemination. ------------------------------ note 1: GRAPHIR-- IR SPECTRAL INTERPRETATION This program takes a list of infrared peak positions and intensities and a list of group frequency assignments. The program then attempts to maximally assign peaks to group frequencies a default assignment table (groups.grp) is read. options are: -h...this message -v show version number -s generate statistics on edge weights in graph and write a histogram file name.hst or name.HST based on the peak file name or the default graph output file name -g name...specify new group table -p name...specify the name of an input file of peak data -f double...specify the value used in the trapezoid weights ]0,.5[ -t double...specify the value used as a threshold for weights => 0 -d name...specify the name of an input file of words that id a group for omission from the graph -r name...specify the name of an input file of words that id a group for restoring an omitted group to the graph -o name...specify file to save graph (defaults to GRAPHED.G) ------------------------------ note 2: A Library of Efficient Data Types and Algorithms Stefan N"aher Max-Planck-Institut f"ur Informatik Im Stadtwald, 6600 Saarbr"ucken, FRG stefan@mpi-sb.mpg.de this is also available by ftp ------------------------------ note 3: SAMPLE consider a file x.pk with contents: 3338 4.71053 3196 2.44185 3004 6.90776 ... stuff deleted to save space 704 3.12357 526 1.00786 and a table of group frequencies in a file groups.grp with contents: 3700 3600 OH gas dilute 3520 3320 NH2 amines amide 3420 3250 OH solid liquid 3360 3340 NH2 primary amide 3320 3250 OH oxime .... stuff deleted 2750 2350 NH3+ amine hydrohalide 2720 2560 P-OH phosphorous oxy acid 2600 2540 SH alkyl mercaptan 2410 2280 PH phosphine .... stuff deleted 520 430 C-O-C ether 510 400 C-N-C amine 400 465 napthalene 440 420 Cl-C=O acid chloride 405 400 S-C...N thiocyanate and another file called kill with contents: S P F ene yne nitro = C-O-C COO carbox Cl then the command (Unix) graphir -p x.pk -d kill would produce: GRAPHIR-- IR SPECTRAL INTERPRETATION Peaks read from x.pk; 20 peaks were read. These peaks were: (3338,4.71053) (3196,2.44185) (3004,6.90776) (2908,4.71053) (2786,1.42712) ... stuff deleted ... (526,1.00786) 171 Group frequencies were read from groups.grp Window fraction used = 0.01 Weight threshold used = 0.001 There are 18 peak nodes and 31 group nodes in the graph. 0.22 sec used score = 3.747748E+05/(3.803040E+05 : 7.681756E+05) all patterns matched the graph was saved in GRAPHED.G 3338.0 best matches OH solid liquid [28263.2] 3196.0 best matches NH2 primary amide [24418.5] 3004.0 best matches NH3+ amino acid [27631.0] 2908.0 best matches aliphatic CH3 CH2 [28263.2] 2786.0 best matches -O-CH3 -N-CH3 [8562.72] 1588.0 best matches NH2 in amino acid [23434.1] (and NH2 in primary alkyl amide ) [23434.1] 1478.0 best matches NH in secondary amide [25751.0] 1498.0 best matches NH3+ in amino acid or hydrochloride [38632.3] (and aromatic ) [38632.3] 1452.0 best matches CH2 aliphatic [24534.1] (and CH3 aliphatic ) [24534.1] 1388.0 best matches t-butyl [14188.2] 1200.0 best matches CN aromatic amine [10774.6] (and C-C-N amine ) [10774.6] 1088.0 best matches C-O-H secondary, tertiary alcohol [11147.4] 1056.0 best matches C-NH2 primary aliphatic amine [22117.0] (and CH-O-H cyclic alcohol ) [22117.0] (and CH2-O-H primary alcohol ) [22117.0] 1020.0 best matches CH-O-H cyclic alcohol [17443.0] (and carbon ring breathing, cyclic ) [17443.0] 1030.0 best matches CH2-O-H primary alcohol [35404.6] 992.0 best matches carbon ring breathing, cyclic [9421.28] 704.0 best matches Ar-OH phenol [18741.4] 526.0 best matches C-Br [6047.16] (and C-I ) [6047.16] (and cycloalkane ring ) [6047.16] (and C n H 2n+1 alkyl ) [6047.16] ... in the above matches in () are all equal to the preceding match for that peak. Score weights are shown in []. Note it took 0.22 seconds to make the assignments. In reaching the assignment chemical information was used from the file kill: S P F ene yne nitro = C-O-C COO carbox Cl telling the code that these keywords flag groups that are known a prior to be unsuitable. By the way x.pk is a list of peaks for phenylproanolamine hydrochloride. ----------------------------------------------------------------------------- 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"