From chemistry-request@server.ccl.net Thu Dec 7 00:38:17 2000 Received: from mailgw.ust.hk (mailgw.ust.hk [143.89.14.35]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id AAA04682 for ; Thu, 7 Dec 2000 00:38:16 -0500 Received: from uxmail.ust.hk (uxmail.ust.hk [143.89.14.30]) by mailgw.ust.hk (8.11.0/8.11.0) with ESMTP id eB75ZVA15503 for ; Thu, 7 Dec 2000 13:35:31 +0800 (HKT) Received: from ust.hk (chz107.ust.hk [143.89.53.152]) by uxmail.ust.hk (8.11.0/8.11.0) with ESMTP id eB75bnX20676 for ; Thu, 7 Dec 2000 13:37:49 +0800 (HKT) Message-ID: <3A2F2275.BB3F216@ust.hk> Date: Thu, 07 Dec 2000 13:39:02 +0800 From: Zhenyang Lin X-Mailer: Mozilla 4.73 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: chemistry@ccl.net Subject: Optimization with a background charge distribution Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Dear Colleagues, Does anyone have experiences with G98 in running geometry optimization calculations in which a background charge distribution is included. Clearly, keyword "charge" can be used for the purpose. During the optimization process, we know that the atom positions in the studied molecule is being changed from one step to others. However, I am not sure how the positions of the input charges change with the geometry during the process. I failed to get a converged geometry when I ran a calculation on a system in which a background charge distribution is included. I guess that while the atom positions are changed during the optimization the charge positions somehow are not changed accordingly. I appreciate any help or suggestions from all of you. Yours, Z Lin From chemistry-request@server.ccl.net Thu Dec 7 05:56:38 2000 Received: from gip.u-picardie.fr (gip.u-picardie.fr [193.49.184.17]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id FAA06168 for ; Thu, 7 Dec 2000 05:56:38 -0500 Received: from labo5 (labo5.sa.u-picardie.fr [195.83.150.83]) by gip.u-picardie.fr (8.9.3/8.9.3) with SMTP id LAA13488 for ; Thu, 7 Dec 2000 11:56:28 +0100 Date: Thu, 7 Dec 2000 11:56:28 +0100 Message-Id: <200012071056.LAA13488@gip.u-picardie.fr> X-Sender: fyd-phar@mailx.u-picardie.fr X-Mailer: Windows Eudora Light Version 1.5.2 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: chemistry@ccl.net From: Francois Dupradeau Subject: Pka value Dear All, Does someone has an idea of the PKa value for the following acid/base : CH3SO3(-)/CH2(-)SO3(-) and CH3SO3Me/CH2(-)SO3Me Thanks Francois -- F.-Y. Dupradeau http://www.u-picardie.fr/UPIC/UPJV/recherche/labos/bpd/fyd.htm From chemistry-request@server.ccl.net Thu Dec 7 06:01:34 2000 Received: from nottingham.ac.uk (pat.ccc.nottingham.ac.uk [128.243.40.194]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id GAA06236 for ; Thu, 7 Dec 2000 06:01:22 -0500 Received: from gotham.ccc.nottingham.ac.uk ([128.243.40.48] helo=unix.ccc.nottingham.ac.uk) by nottingham.ac.uk with esmtp (Exim 3.16 #1) id 143ymY-0006UF-00 for chemistry@ccl.net; Thu, 07 Dec 2000 11:00:10 +0000 Received: from granby ([128.243.40.43] helo=granby.ccc.nottingham.ac.uk) by unix.ccc.nottingham.ac.uk with esmtp (Exim 3.16 #1) id 143ynS-0004y4-00 for chemistry@ccl.net; Thu, 07 Dec 2000 11:01:06 +0000 Date: Thu, 7 Dec 2000 11:01:05 +0000 (GMT) From: Simon Cross X-Sender: pcxsc@granby.ccc.nottingham.ac.uk To: ccl list Subject: minimisation output Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Does anyone know if there's a way to output coords from each step in an Amber5 minimisation, in a similar way to MD, as ntwr and ntwx seem to be MD specific. Thanks in advance, ----------------------------------------- Simon Cross School of Chemistry University of Nottingham tel. 0115 9514193 Email: pcxsc@nottingham.ac.uk From chemistry-request@server.ccl.net Thu Dec 7 06:52:58 2000 Received: from mailgw.barc.ernet.in ([202.54.18.131]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id GAA07197 for ; Thu, 7 Dec 2000 06:52:52 -0500 Received: from apsara.barc.ernet.in (apsara.barc.ernet.in [192.168.1.21]) by mailgw.barc.ernet.in (8.9.3/8.9.3) with ESMTP id RAA12419 for ; Thu, 7 Dec 2000 17:29:42 +0530 (IST) (envelope-from tapang@apsara.barc.ernet.in) Received: from localhost (tapang@localhost) by apsara.barc.ernet.in (8.9.3/8.9.3) with ESMTP id RAA19158 for ; Thu, 7 Dec 2000 17:28:07 +0530 Date: Thu, 7 Dec 2000 17:28:07 +0530 (IST) From: "Dr.Tapan K.Ghanty" To: chemistry@ccl.net Subject: Iodine basis set problem Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi Friends, While doing molecular HF calculations using 6-311G** basis for Iodine, (the iodine basis set got from emsl.pnl.gov website) I got the following warnings. I am using GAMESS-US program. I do not know the exact reason for getting these messages. Does it mean that the 6-311G** basis set is not properly optimized or rather not properly normalized? Any suggestion? Is there any other comparable basis sets for iodine? I know about DZVP basis, optimized for DFT wavefunctions by Salahub and co-workers. *** WARNING! ATOM 15 SHELL 71 TYPE S HAS NORMALIZATION 2.09704269 *** WARNING! ATOM 15 SHELL 72 TYPE S HAS NORMALIZATION 1.65864942 *** WARNING! ATOM 15 SHELL 81 TYPE P HAS NORMALIZATION 1.28533575 *** WARNING! ATOM 15 SHELL 90 TYPE D HAS NORMALIZATION 1.60702361 Also, is there any option in GAMESS-US program to select 5-components of d functions instead of 6-components. I hope this will solve my second (following) problem. Any other suggestions? * * * WARNING * * * 2 EIGENVALUES OF THE OVERLAP ARE BELOW 1.0E-05 OF WHICH THE SMALLEST IS 3.46345E-07 THIS MAY INDICATE PARTIAL LINEAR DEPENDENCE OF YOUR BASIS. Following is the iodine basis set used in my calculation -------------------------------------------------------- IODINE ! (15s,12p,6d) -> [10s,9p,4d] S 5 1 444750.0000 0.8900000000E-03 2 66127.00000 0.6940000000E-02 3 14815.00000 0.3609000000E-01 4 4144.900000 0.1356800000 5 1361.200000 0.3387800000 S 2 1 508.4400000 0.4365900000 2 209.5900000 0.1837500000 S 1 1 81.95900000 1.000000000 S 1 1 36.80500000 1.000000000 S 1 1 13.49500000 1.000000000 S 1 1 6.885900000 1.000000000 S 1 1 2.552000000 1.000000000 S 1 1 1.208800000 1.000000000 S 1 1 0.2734000000 1.000000000 S 1 1 0.1009000000 1.000000000 P 4 1 2953.600000 0.1221000000E-01 2 712.6100000 0.8587000000E-01 3 236.7100000 0.2949300000 4 92.63100000 0.4784900000 P 1 1 39.73200000 1.000000000 P 1 1 17.27300000 1.000000000 P 1 1 7.957000000 1.000000000 P 1 1 3.152900000 1.000000000 P 1 1 1.332800000 1.000000000 P 1 1 0.4947000000 1.000000000 P 1 1 0.2160000000 1.000000000 P 1 1 0.8293000000E-01 1.000000000 D 3 1 261.9500000 0.3144000000E-01 2 76.73400000 0.1902800000 3 27.55100000 0.4724700000 D 1 1 10.60600000 1.000000000 D 1 1 3.421700000 1.000000000 D 1 1 1.137000000 1.000000000 IODINE ! (1d) D 1 1 0.302000000 1.00000000 Following is the GAMESS-US output file ------------------------------------------------------------------ IODINE 71 S 169 444750.000000 22.908378 ( 0.000890) 71 S 170 66127.000000 42.772137 ( 0.006940) 71 S 171 14815.000000 72.432002 ( 0.036090) 71 S 172 4144.900000 104.753548 ( 0.135680) 71 S 173 1361.200000 113.468831 ( 0.338780) 72 S 174 508.440000 55.260939 ( 0.436590) 72 S 175 209.590000 11.965200 ( 0.183750) 73 S 176 81.959000 19.413667 ( 1.000000) 74 S 177 36.805000 10.649770 ( 1.000000) 75 S 178 13.495000 5.018105 ( 1.000000) 76 S 179 6.885900 3.029567 ( 1.000000) 77 S 180 2.552000 1.439032 ( 1.000000) 78 S 181 1.208800 0.821629 ( 1.000000) 79 S 182 0.273400 0.269469 ( 1.000000) 80 S 183 0.100900 0.127593 ( 1.000000) 81 P 184 2953.600000 487.093054 ( 0.012210) 81 P 185 712.610000 579.245978 ( 0.085870) 81 P 186 236.710000 501.702634 ( 0.294930) 81 P 187 92.631000 251.927154 ( 0.478490) 82 P 188 39.732000 142.188876 ( 1.000000) 83 P 189 17.273000 50.193769 ( 1.000000) 84 P 190 7.957000 19.049202 ( 1.000000) 85 P 191 3.152900 5.988635 ( 1.000000) 86 P 192 1.332800 2.041251 ( 1.000000) 87 P 193 0.494700 0.591381 ( 1.000000) 88 P 194 0.216000 0.209898 ( 1.000000) 89 P 195 0.082930 0.063435 ( 1.000000) 90 D 196 261.950000 1418.392596 ( 0.031440) 90 D 197 76.734000 1001.275460 ( 0.190280) 90 D 198 27.551000 414.043854 ( 0.472470) 91 D 199 10.606000 102.594669 ( 1.000000) 92 D 200 3.421700 14.168792 ( 1.000000) 93 D 201 1.137000 2.060584 ( 1.000000) 94 D 202 0.302000 0.202498 ( 1.000000) With thanks, Dr. Tapan K. Ghanty Theoretical Chemistry Section Chemistry Division Bhabha Atomic Research Centre Trombay, Mumbai 400 085 India From chemistry-request@server.ccl.net Thu Dec 7 10:34:38 2000 Received: from hamiltonian.chem.cornell.edu (root@hamiltonian.chem.cornell.edu [128.253.229.83]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id KAA08333 for ; Thu, 7 Dec 2000 10:34:38 -0500 From: marketa@hamiltonian.chem.cornell.edu Received: from localhost (marketa@localhost) by hamiltonian.chem.cornell.edu (8.11.0/8.11.0) with ESMTP id eB7GaPx30793 for ; Thu, 7 Dec 2000 10:36:25 -0600 Date: Thu, 7 Dec 2000 10:36:25 -0600 (CST) To: chemistry@ccl.net Subject: Summary: C60 coordinates Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi, it turned out that many people are interested in getting the replies to my ccl quesiton > Hi, > I wonder if someone could provide me with a Z-matrix input > for buckminster-fullerene C60. I need my program to recognize > as much of the icosahedral symmetry as possible. > > Thank you, > Marketa Munzarova So, here are the replies: Gert von Helden's sent me a very elegant Z-matrix input, based on building the pentagons and hexagons on C60 surface: **************************************************************************** Hi, G98 should recognize the full symmetry of C60 with the following z-matrix (note that two angles, instead of one dihedral, is used for atom 6): Hope that helps, Gert ----------------------- # AM1 opt C60 0 1 c c 1 r5 c 2 r5 1 108.0 c 3 r5 2 108.0 1 0.00 c 4 r5 3 108.0 2 0.00 c 5 r6 4 120.0 1 120.0 1 c 6 r5 5 120.0 4 0.00 c 7 r6 6 120.0 5 0.00 c 8 r5 7 120.0 6 0.00 c 9 r5 4 120.0 3 0.00 c 10 r6 9 120.0 4 0.00 c 11 r5 10 120.0 9 0.00 c 12 r5 3 120.0 2 0.00 c 13 r6 12 120.0 3 0.00 c 14 r5 13 120.0 2 0.00 c 15 r5 2 120.0 1 0.00 c 16 r5 15 108.0 14 0.00 c 17 r5 16 108.0 15 0.00 c 18 r6 14 120.0 13 0.00 c 19 r5 18 120.0 14 0.00 c 20 r6 19 120.0 13 0.00 c 21 r5 13 108.0 12 0.00 c 22 r6 11 120.0 10 0.00 c 23 r5 22 120.0 11 0.00 c 10 r5 11 120.0 22 0.00 c 11 r5 10 120.0 9 0.00 c 12 r5 3 120.0 2 0.00 c 13 r6 12 120.0 3 0.00 c 14 r5 13 120.0 2 0.00 c 15 r5 2 120.0 1 0.00 c 16 r5 15 108.0 14 0.00 c 17 r5 16 108.0 15 0.00 c 18 r6 14 120.0 13 0.00 c 19 r5 18 120.0 14 0.00 c 20 r6 19 120.0 13 0.00 c 21 r5 13 108.0 12 0.00 c 22 r6 11 120.0 10 0.00 c 23 r5 22 120.0 11 0.00 c 10 r5 11 120.0 22 0.00 c 8 r5 9 108.0 10 0.00 c 26 r6 8 120.0 7 0.00 c 27 r5 26 120.0 8 0.00 c 28 r6 27 120.0 26 0.00 c 29 r5 7 108.0 6 0.00 c 30 r6 29 120.0 28 0.00 c 31 r5 30 120.0 29 0.00 c 32 r6 31 120.0 30 0.00 c 33 r5 28 108.0 27 0.00 c 34 r5 33 108.0 28 0.00 c 35 r6 27 120.0 26 0.00 c 36 r5 24 108.0 23 0.00 c 37 r5 36 108.0 24 0.00 c 38 r6 23 120.0 22 0.00 c 39 r5 38 120.0 37 0.00 c 40 r5 39 108.0 20 0.00 c 41 r6 19 120.0 18 0.00 c 42 r5 41 120.0 19 0.00 c 43 r5 17 120.0 16 0.00 c 44 r5 43 108.0 42 0.00 c 45 r5 44 108.0 43 0.00 c 46 r6 42 120.0 41 0.00 c 47 r5 46 120.0 42 0.00 c 48 r5 40 120.0 39 0.00 c 49 r5 48 108.0 47 0.00 c 50 r5 49 108.0 48 0.00 c 51 r6 47 120.0 46 0.00 c 52 r5 51 120.0 47 0.00 c 53 r5 52 108.0 32 0.00 c 54 r6 53 120.0 45 0.00 c 55 r5 54 120.0 53 0.00 c 56 r5 44 120.0 43 0.00 c 57 r5 56 108.0 55 0.00 c 58 r5 57 108.0 56 0.00 c 59 r6 55 120.0 54 0.00 r5 1.474 r6 1.420 *********************************************************************************** Another very nice Z-matrix input, definig the carbons with respect to the C60 centre and a number of dummies, has been sent by Roy Jensen ----------------------------------------------------------------------------------- Gaussian test jobs 322 and 333 have C60 in z-matrix format. I have copied 322 below if you do not have it. Roy Jensen ------------------- %chk=test322 %mem=2000000,4000000 #p RHF/STO-3G test opt freq Gaussian Test Job 322 (Part 1): C60 Icosahedral 0 1 X X 1 1. X 1 1. 2 90. X 1 1. 2 90. 3 180. X 1 1. 3 C1 2 0. X 1 1. 3 C1 5 C2 X 1 1. 3 C1 5 C3 X 1 1. 3 C1 5 -C2 X 1 1. 3 C1 5 -C3 X 1 1. 4 C1 2 180. X 1 1. 4 C1 10 C2 X 1 1. 4 C1 10 C3 X 1 1. 4 C1 10 -C2 X 1 1. 4 C1 10 -C3 C 1 R 3 A 5 0. C 1 R 3 A 6 0. C 1 R 3 A 7 0. C 1 R 3 A 8 0. C 1 R 3 A 9 0. C 1 R 5 A 3 0. C 1 R 5 A 3 C2 C 1 R 5 A 3 C3 C 1 R 5 A 3 -C2 C 1 R 5 A 3 -C3 C 1 R 6 A 3 0. C 1 R 6 A 3 C2 C 1 R 6 A 3 C3 C 1 R 6 A 3 -C2 C 1 R 6 A 3 -C3 C 1 R 7 A 3 0. C 1 R 7 A 3 C2 C 1 R 7 A 3 C3 C 1 R 7 A 3 -C2 C 1 R 7 A 3 -C3 C 1 R 8 A 3 0. C 1 R 8 A 3 C2 C 1 R 8 A 3 C3 C 1 R 8 A 3 -C2 C 1 R 8 A 3 -C3 C 1 R 9 A 3 0. C 1 R 9 A 3 C2 C 1 R 9 A 3 C3 C 1 R 9 A 3 -C2 C 1 R 9 A 3 -C3 C 1 R 4 A 10 0. C 1 R 4 A 11 0. C 1 R 4 A 12 0. C 1 R 4 A 13 0. C 1 R 4 A 14 0. C 1 R 10 A 4 0. C 1 R 10 A 4 C2 C 1 R 10 A 4 C3 C 1 R 10 A 4 -C2 C 1 R 10 A 4 -C3 C 1 R 11 A 4 0. C 1 R 11 A 4 C2 C 1 R 11 A 4 C3 C 1 R 11 A 4 -C2 C 1 R 11 A 4 -C3 C 1 R 12 A 4 0. C 1 R 12 A 4 C2 C 1 R 12 A 4 C3 C 1 R 12 A 4 -C2 C 1 R 12 A 4 -C3 C 1 R 13 A 4 0. C 1 R 13 A 4 C2 C 1 R 13 A 4 C3 C 1 R 13 A 4 -C2 C 1 R 13 A 4 -C3 C 1 R 14 A 4 0. C 1 R 14 A 4 C2 C 1 R 14 A 4 C3 C 1 R 14 A 4 -C2 C 1 R 14 A 4 -C3 Variables: R 3.52429 A 20.5346 Constants: C1 63.434948823 C2 72.0 C3 144.0 ******************************************************************************** Several people sent me cartesian coordinates inputs, which can be converted to full Z-matrixes. Here are the cartesian coordinates from N. Dragoe, in G98 format: ******************************************************************************** Hi, These are the coordinates for C60. I am not sure but I think you could transfer them into Z-matrix with Gaussian. -------- Standard orientation: --------------------------------------------------------------------- Center Atomic Atomic Coordinates (Angstroms) Number Number Type X Y Z --------------------------------------------------------------------- 1 6 0 0.726656 -1.000157 3.300459 2 6 0 1.175755 0.382026 3.300459 3 6 0 1.410183 -1.940951 2.581756 4 6 0 2.281725 0.741377 2.581756 5 6 0 2.281725 1.977639 1.817704 6 6 0 0.000000 1.236262 3.300459 7 6 0 0.000000 2.399147 2.581756 8 6 0 1.175755 2.781173 1.817704 9 6 0 -0.683527 -2.941108 1.817704 10 6 0 -1.410183 -1.940951 2.581756 11 6 0 0.683527 -2.941108 1.817704 12 6 0 -0.726656 -1.000157 3.300459 13 6 0 -1.175755 0.382026 3.300459 14 6 0 -2.585938 -1.558925 1.817704 15 6 0 -3.008381 -0.258779 1.817704 16 6 0 -2.281725 0.741377 2.581756 17 6 0 0.726656 -3.399304 -0.581443 18 6 0 -0.726656 -3.399304 -0.581443 19 6 0 1.410183 -3.177213 0.581443 20 6 0 -1.410183 -3.177213 0.581443 21 6 0 -2.585938 -2.322977 0.581443 22 6 0 -1.175755 -2.781173 -1.817704 23 6 0 -2.281725 -1.977639 -1.817704 24 6 0 -3.008381 -1.741534 -0.581443 25 6 0 3.008381 -1.741534 -0.581443 26 6 0 2.281725 -1.977639 -1.817704 27 6 0 2.585938 -2.322977 0.581443 28 6 0 1.175755 -2.781173 -1.817704 29 6 0 0.000000 -2.399147 -2.581756 30 6 0 2.281725 -0.741377 -2.581756 31 6 0 1.175755 -0.382026 -3.300459 32 6 0 0.000000 -1.236262 -3.300459 33 6 0 3.008381 -0.258779 1.817704 34 6 0 3.457479 0.359351 0.581443 35 6 0 2.585938 -1.558925 1.817704 36 6 0 3.457479 -0.359351 -0.581443 37 6 0 3.008381 0.258779 -1.817704 38 6 0 3.008381 1.741534 0.581443 39 6 0 2.585938 2.322977 -0.581443 40 6 0 2.585938 1.558925 -1.817704 41 6 0 -0.726656 1.000157 -3.300459 42 6 0 0.726656 1.000157 -3.300459 43 6 0 1.410183 1.940951 -2.581756 44 6 0 -1.410183 1.940951 -2.581756 45 6 0 -3.008381 0.258779 -1.817704 46 6 0 -2.281725 -0.741377 -2.581756 47 6 0 -1.175755 -0.382026 -3.300459 48 6 0 -2.585938 1.558925 -1.817704 49 6 0 -3.008381 1.741534 0.581443 50 6 0 -3.457479 0.359351 0.581443 51 6 0 -3.457479 -0.359351 -0.581443 52 6 0 -2.585938 2.322977 -0.581443 53 6 0 -0.726656 3.399304 0.581443 54 6 0 -1.175755 2.781173 1.817704 55 6 0 -2.281725 1.977639 1.817704 56 6 0 -1.410183 3.177213 -0.581443 57 6 0 0.683527 2.941108 -1.817704 58 6 0 1.410183 3.177213 -0.581443 59 6 0 0.726656 3.399304 0.581443 60 6 0 -0.683527 2.941108 -1.817704 ****************************************************************************************** James Stewart provided me with cartesian coordinates in MOPAC format: ------------------------------------------------------------------------ c60.MOP C 0.000000 0 0.000000 0 0.000000 0 0 0 0 C 1.457458 1 0.000000 0 0.000000 0 1 0 0 C 1.457489 1 107.999603 1 0.000000 0 2 1 0 C 1.457474 1 107.999603 1 0.000000 1 3 2 1 C 1.457443 1 108.001343 1 0.000000 1 1 2 3 C 5.199631 1 90.000244 1 31.718307 1 1 2 3 C 1.457489 1 35.998703 1 238.281448 1 6 1 2 C 1.383911 1 119.999756 1 217.376038 1 4 3 2 C 1.457474 1 120.001495 1 0.000000 1 8 4 3 C 1.457458 1 108.001343 1 0.000000 1 6 7 8 C 2.841415 1 59.999008 1 142.622208 1 1 2 3 C 1.457474 1 144.000046 1 280.812653 1 11 1 2 C 1.457458 1 107.999603 1 63.433105 1 12 11 1 C 1.457489 1 107.999603 1 0.000000 1 13 12 11 C 1.383926 1 120.001495 1 217.379532 1 2 1 5 C 6.651367 1 77.666138 1 52.436615 1 1 2 3 C 1.383926 1 120.001495 1 142.622208 1 10 6 7 C 1.457489 1 120.001495 1 221.812057 1 17 10 6 C 1.457489 1 107.999603 1 142.622208 1 18 17 10 C 1.457474 1 107.999603 1 217.376038 1 16 17 10 C 1.383926 1 119.999756 1 359.683517 1 16 17 10 C 1.457474 1 119.999756 1 221.810303 1 21 16 17 C 1.457489 1 107.999603 1 142.622208 1 22 21 16 C 1.457504 1 107.999603 1 0.000000 1 23 22 21 C 1.383926 1 119.999756 1 142.622208 1 6 7 8 C 1.383942 1 119.998000 1 142.623962 1 1 2 3 C 1.457489 1 119.999756 1 221.810303 1 26 1 2 C 1.457489 1 107.999603 1 142.623962 1 27 26 1 C 1.457504 1 107.999603 1 0.000000 1 28 27 26 C 1.383926 1 119.999756 1 142.622208 1 11 12 13 C 1.383911 1 119.999756 1 217.376038 1 3 2 1 C 1.383896 1 119.999756 1 217.376038 1 14 13 12 C 1.457489 1 119.998000 1 0.000000 1 32 14 13 C 1.383911 1 119.999756 1 0.000000 1 18 17 10 C 1.383926 1 119.998000 1 0.000000 1 9 8 4 C 1.383926 1 119.999756 1 142.623962 1 5 1 2 C 1.383926 1 119.999756 1 217.377792 1 7 6 10 C 1.383911 1 119.999756 1 217.376038 1 24 23 22 C 1.457489 1 119.999756 1 0.000000 1 38 24 23 C 1.383881 1 119.999756 1 0.000000 1 27 26 1 C 1.383926 1 120.001495 1 217.379532 1 23 22 21 C 1.457474 1 119.999756 1 0.000000 1 41 23 22 C 1.457474 1 107.999603 1 142.623962 1 42 41 23 C 1.383865 1 119.999756 1 217.376038 1 28 27 26 C 1.383881 1 120.001495 1 0.000000 1 39 38 24 C 1.457474 1 119.999756 1 0.000000 1 41 23 22 C 1.457474 1 107.999603 1 142.623962 1 42 41 23 C 1.383865 1 119.999756 1 217.376038 1 28 27 26 C 1.383881 1 120.001495 1 0.000000 1 39 38 24 C 6.054962 1 60.000748 1 100.813156 1 1 2 3 C 1.457474 1 107.999603 1 259.186844 1 46 1 2 C 1.383896 1 119.999756 1 217.377792 1 19 18 17 C 1.383881 1 120.001495 1 0.000000 1 33 32 14 C 1.383896 1 120.001495 1 217.376038 1 13 12 11 C 1.383911 1 119.999756 1 217.376038 1 43 42 41 C 1.457489 1 119.998000 1 0.000000 1 51 43 42 C 1.383926 1 119.999756 1 142.623962 1 46 47 48 C 1.383911 1 120.001495 1 217.376038 1 12 11 15 C 1.383911 1 120.001495 1 217.377792 1 29 28 27 C 1.383926 1 119.999756 1 142.622208 1 20 16 17 C 1.383911 1 119.998000 1 217.376038 1 47 46 50 C 1.383896 1 120.001495 1 0.000000 1 52 51 43 C 1.383926 1 119.999756 1 0.000000 1 42 41 23 C 1.383911 1 119.999756 1 0.000000 1 22 21 16 *********************************************************************************** Pascal Bonnet sent me both cartesian coordinates and their conversion to the full Z-matrix. Either can be obtained at the web page of prof. Yoshida: http://shachi.cochem2.tutkie.tut.ac.jp/Fuller/Fuller.html *********************************************************************************** Geoff Hutchison's reply included a full Z-matrix input: *********************************************************************************** As it happens, I had a BuckyBall Molfile, so I converted it to Gaussian Z-matrix for you. See the attached file. Cheers, -Geoff Hutchison Northwestern Chemistry ------------------------ C C 1 r2 C 2 r3 1 a3 C 3 r4 2 a4 1 d4 C 4 r5 3 a5 2 d5 C 1 r6 2 a6 3 d6 C 4 r7 3 a7 2 d7 C 7 r8 4 a8 3 d8 C 8 r9 7 a9 4 d9 C 7 r10 4 a10 3 d10 C 10 r11 7 a11 4 d11 C 11 r12 10 a12 7 d12 C 12 r13 11 a13 10 d13 C 10 r14 7 a14 4 d14 C 3 r15 2 a15 1 d15 C 13 r16 12 a16 11 d16 C 16 r17 13 a17 12 d17 C 17 r18 16 a18 13 d18 C 18 r19 17 a19 16 d19 C 6 r20 1 a20 2 d20 C 1 r21 2 a21 3 d21 C 21 r22 1 a22 2 d22 C 19 r23 18 a23 17 d23 C 23 r24 19 a24 18 d24 C 16 r25 13 a25 12 d25 C 12 r26 11 a26 10 d26 C 11 r27 10 a27 7 d27 C 27 r28 11 a28 10 d28 C 15 r29 3 a29 2 d29 C 29 r30 15 a30 3 d30 C 30 r31 29 a31 15 d31 C 31 r32 30 a32 29 d32 C 22 r33 21 a33 1 d33 C 33 r34 22 a34 21 d34 C 24 r35 23 a35 19 d35 C 35 r36 24 a36 23 d36 C 26 r37 12 a37 11 d37 C 37 r38 26 a38 12 d38 C 28 r39 27 a39 11 d39 C 39 r40 28 a40 27 d40 C 40 r41 39 a41 28 d41 C 31 r42 30 a42 29 d42 C 32 r43 31 a43 30 d43 C 33 r44 22 a44 21 d44 C 34 r45 33 a45 22 d45 C 35 r46 24 a46 23 d46 C 36 r47 35 a47 24 d47 C 37 r48 26 a48 12 d48 C 38 r49 37 a49 26 d49 C 39 r50 28 a50 27 d50 C 41 r51 40 a51 39 d51 C 42 r52 31 a52 30 d52 C 44 r53 33 a53 22 d53 C 46 r54 35 a54 24 d54 C 49 r55 38 a55 37 d55 C 55 r56 49 a56 38 d56 C 56 r57 55 a57 49 d57 C 57 r58 56 a58 55 d58 C 58 r59 57 a59 56 d59 C 59 r60 58 a60 57 d60 Variables: r2= 1.3802 r3= 1.3706 a3= 120.01 r4= 1.3802 a4= 120.00 d4= 359.97 r5= 1.3706 a5= 119.99 d5= 0.03 r6= 1.3705 a6= 119.99 d6= 0.03 r7= 1.3704 a7= 119.96 d7= 138.15 r8= 1.3704 a8= 107.98 d8= 217.44 r9= 1.3703 a9= 108.02 d9= 359.97 r10= 1.3804 a10= 120.07 d10= 359.97 r11= 1.3715 a11= 120.11 d11= 221.69 r12= 1.3799 a12= 119.87 d12= 0.15 r13= 1.3698 a13= 120.05 d13= 0.03 r14= 1.3704 a14= 119.97 d14= 0.03 r15= 1.3702 a15= 108.01 d15= 142.57 r16= 1.3696 a16= 108.08 d16= 142.73 r17= 1.3799 a17= 120.04 d17= 217.29 r18= 1.3716 a18= 119.89 d18= 359.97 r19= 1.3705 a19= 108.04 d19= 142.58 r20= 1.3703 a20= 107.99 d20= 217.44 r21= 1.3704 a21= 119.97 d21= 221.85 r22= 1.3702 a22= 107.97 d22= 142.56 r23= 1.3715 a23= 108.02 d23= 359.97 r24= 1.3693 a24= 107.83 d24= 0.03 r25= 1.3736 a25= 108.08 d25= 0.20 r26= 1.3729 a26= 120.09 d26= 138.56 r27= 1.3694 a27= 107.85 d27= 142.75 r28= 1.3693 a28= 108.26 d28= 359.97 r29= 1.3703 a29= 107.97 d29= 359.97 r30= 1.3702 a30= 108.04 d30= 0.03 r31= 1.3804 a31= 119.93 d31= 217.36 r32= 1.3704 a32= 119.98 d32= 138.16 r33= 1.3804 a33= 120.00 d33= 217.33 r34= 1.3695 a34= 120.01 d34= 138.37 r35= 1.3848 a35= 120.01 d35= 216.94 r36= 1.3739 a36= 119.83 d36= 139.17 r37= 1.3852 a37= 119.80 d37= 0.46 r38= 1.3746 a38= 120.08 d38= 358.59 r39= 1.3800 a39= 120.21 d39= 217.58 r40= 1.3696 a40= 120.04 d40= 137.97 r41= 1.3696 a41= 108.10 d41= 217.28 r42= 1.3715 a42= 120.10 d42= 359.81 r43= 1.3716 a43= 108.03 d43= 217.19 r44= 1.3697 a44= 120.08 d44= 359.89 r45= 1.3731 a45= 108.06 d45= 216.90 r46= 1.3746 a46= 119.80 d46= 359.63 r47= 1.3426 a47= 108.01 d47= 217.28 r48= 1.3426 a48= 119.25 d48= 221.15 r49= 1.3740 a49= 108.98 d49= 217.92 r50= 1.3735 a50= 120.08 d50= 359.46 r51= 1.3728 a51= 108.06 d51= 359.79 r52= 1.3693 a52= 107.86 d52= 142.77 r53= 1.3735 a53= 108.09 d53= 143.06 r54= 1.3425 a54= 108.13 d54= 142.98 r55= 1.3425 a55= 108.04 d55= 0.71 r56= 1.4358 a56= 120.29 d56= 141.94 r57= 1.3426 a57= 120.10 d57= 359.48 r58= 1.3747 a58= 108.15 d58= 218.50 r59= 1.3741 a59= 108.98 d59= 359.54 r60= 1.3427 a60= 108.02 d60= 0.71 ******************************************************************************************** Finally, here is Marcel Swart's advice: ------------- Try building it, save it in Cartesian coordinates and afterwards use Babel to do the conversion to Z-matrix. Marcel Swart ******************************************************************************************** Many thanks to everybody who replied! Marketa ****************************************************************** Marketa L. Munzarova e-mail: mm335@cornell.edu 220 Baker Lab phone: 1-607-255-0597 Cornell University Ithaca, New York, 14850-1301 ****************************************************************** From chemistry-request@server.ccl.net Thu Dec 7 14:58:08 2000 Received: from ccl.net (atlantis.ccl.net [192.148.249.4]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id OAA09399 for ; Thu, 7 Dec 2000 14:58:08 -0500 Received: from oscdesk53.ccl.net (oscdesk53.ccl.net [192.148.249.53]) by ccl.net (8.9.3/8.9.3/OSC 2.0) with ESMTP id OAA15051; Thu, 7 Dec 2000 14:58:08 -0500 (EST) Date: Thu, 7 Dec 2000 14:58:08 -0500 From: Gerald Lushington To: chemistry@ccl.net cc: gerald@ccl.net Subject: Re: CCL:G98 / displacement issue with frozen coordinates In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Greetings once again. I also received a helpful note from Frank Jensen in response to my original query. He also sent a program which level-shifts away the effects of the frozen atoms which I think is an effective means for addressing my point 1). Anyway, his note is appended below. Thanks again to all who responded! - Gerry Forwarded note: Date: Fri, 1 Dec 2000 07:49:46 +0100 (MET) From: Frank Jensen To: Gerald Lushington Subject: Re: CCL:G98 / displacement issue with frozen coordinates Gerald, I have encountered the same problem some time ago, it is clearly a bug in G98. I reported it to Gaussian, and they put it on the 'fix'-list. This was with rev. A7, I don't know if it has been fixed in rev. A9. Anyway, once the forces drop below the cutoff, just terminate the run. For frequencies only the residual forces matter. And on this note, if you need to get frequencies for constrained optimized structure, I have a small program for level-shifting the frozen atoms away before determining the frequencies. That way you get 'real' frequencies within the constraints that you have put on the system, i.e. all contributions from frozen atoms are completely decoupled from the remaining degrees of freedom. If you are interested drop me a note. Frank -------------------------------------------------- | Frank Jensen, Department of Chemistry | | SDU Odense University, DK-5230 Odense, Denmark | | FAX +45 66 15 87 80 , Voice +45 65 50 25 07 | | http://bogense.chem.ou.dk/~frj | -------------------------------------------------- On Thu, 30 Nov 2000, Gerald Lushington wrote: > Hi again! > > No intention of closing down this discussion so quickly if others have points > to make, but I thought I'd take a moment and share the three very insightful > responses I've gotten so far. Each response addressed different aspects of the > problem and in each case proved very helpful. Glad to see you're all so > coordinated ;) > > Jeremy Greenwood confirmed that convergence is slow, but also that it's > generally possible to (eventually) achieve. You may wish to refer below to his > response for a few constraining tips he suggests. > > Nicolas Ferre gave a few suggestions as to how to circumvent the strange > "Delta X" and "New X" values we were getting. Turns out that using the "NOSYM" > keyword straightens this glitch out nicely! > > Finally, as you have likely seen, Stefan Fau provided a very thoughtful and > detailed discussion on the validity of frequency analysis on our constrained > system. Have to sit down and think carefully about this, but his suggestions > make sense to me on the first pass. > > Anyway, thanks all! Original question and the three responses appended below. > > - Gerry > > ---------------------------------------------------------------------- > Gerald H. Lushington Ph: 614-292-6036 > Research Specialist Fax: 614-292-7168 > OSC / PET-CCM e-mail: gerald@ccl.net > 1224 Kinnear Road http://www.arl.hpc.mil/PET/cta/ccm > Columbus OH 43212-1163 http://www.asc.hpc.mil/PET/CCM > > > > == Original Question ========================================================== > Dear all, > > In working with a colleague on large cluster optimizations of enzyme active > site (via G98), I've encountered a problem that seems to be: > > - part philosophical > - part methodological > - part bug (?) > > Anyway, the issue has to do with doing geometry optimizations with some degrees > of freedom contrained. In general it's proving necessary to freeze the > coordinates for at least a couple of atoms in order to keep the cluster in some > sort of shape comparable to the real biosystem. The tradeoff, of course, comes > in the fact that forces and displacements will never achieve global minima. > Standard protocol seems to be to run the job until these quantities level off > to some reasonably converged values. When we apply this strategy we find that > force and net displacement eventually level off, but the net displacement > remains very high. This brings me to question number 1: > > 1) If we stop the geometry optimization at a structure whose force and > displacement values have leveled, but whose displacement is still high, > would any subsequent vibrational analysis be at all physically meaningful? > Are there any effective guidelines for this? > > My first thought on this was that Gaussian would likely decompose both force > and displacement as a sum over coordinates. Since the only freezing we do is > way out on the perifery of the cluster, while the structural and vibrational > information of interest is in the center of the system, maybe we can track the > forces and displacements in the central region of interest and feel comfortable > about the results when those central contributions are low. > > So, to try to follow that strategy, we've been refering to a table generated as > standard G98 output during the geometry optimization that looks like: > > Variable Old X -DE/DX Delta X Delta X Delta X New X > (Linear) (Quad) (Total) > X1 -2.50100 -0.00465 0.00000 0.00000 2.50100 0.00000 > Y1 8.90656 0.01749 0.00000 0.00000 -8.90656 0.00000 > Z1 -8.10547 -0.01647 0.00000 0.00000 8.10547 0.00000 > X11 -3.32922 0.00421 0.00000 0.00000 2.50100 -0.82822 > Y11 -9.47808 -0.00542 0.00000 0.00000 -8.90656 -18.38465 > Z11 -1.01768 -0.00913 0.00000 0.00000 8.10547 7.08779 > X38 3.49152 0.00310 0.00000 0.00000 2.50100 5.99252 > Y38 14.83066 0.00184 0.00000 0.00000 -8.90656 5.92410 > Z38 -5.24649 -0.00415 0.00000 0.00000 8.10547 2.85898 > X48 -1.96597 0.00535 0.00000 0.00000 2.50100 0.53502 > Y48 -4.25077 -0.01552 0.00000 0.00000 -8.90656 -13.15733 > Z48 9.58721 0.02164 0.00000 0.00000 8.10547 17.69267 > X50 1.54048 -0.00800 0.00000 0.00000 2.50100 4.04148 > Y50 -8.97461 0.00161 0.00000 0.00000 -8.90656 -17.88117 > Z50 5.17857 0.00810 0.00000 0.00000 8.10547 13.28404 > R1 2.84387 0.01822 0.00000 0.00000 0.00000 2.84388 > R2 2.02201 0.01493 0.00000 0.00000 0.00000 2.02201 > R3 2.02201 0.01403 0.00000 0.00000 0.00000 2.02201 > R4 2.02201 0.01490 0.00000 0.00000 0.00000 2.02201 > R5 2.81303 -0.01054 0.00000 -0.01044 -0.00995 2.80307 > R6 2.10327 -0.02056 0.00000 -0.01706 -0.01706 2.08621 > R7 2.10327 -0.01967 0.00000 -0.01633 -0.01633 2.08694 > R8 2.60262 -0.00578 0.00000 -0.00448 -0.00457 2.59805 > R9 2.55751 -0.00018 0.00000 -0.00230 -0.00185 2.55566 > .. .. .. > > NOTE: in the above, the first bunch of coordinates (X1-R4 inclusive) are all > frozen. R5 onwards (many coordinates not shown) are freely varying. > > Anyway, in the variable coordinates toward the bottom, it seems that the > "DE/DX" is a real force, the "Delta X" are real displacements, and the "Old X" > and "New X" are exactly as they sound -- initial and final values for the > coordinate. However: > > 2) What's going on with these the "Delta X" and "New X" values for the frozen > coordinates?? They're frozen (and really do not vary during the > optimization) so shouldn't "New X" = "Old X" regardless of nonzero forces > and predicted displacements? > > 3) Furthermore, the values written in "Delta X (Total)" look extremely fishy to > me? They're all either "2.50100", "-8.90656", "8.10547" or "0.00000"! Is > this a bug, and if so, might it be corrupting the value given for net > displacement at the end of each geometry step? > > Anyway, this may be a complex issue, but any insight you may have would be > greatly appreciated! > > - Gerry > > > P.S. I did drop Gaussian Inc. a note on this several weeks ago -- I guess it's > somewhere in their queue. > > > == Response #1: Jeremy Greenwood ==================== > > I am doing the same kind of thing at the moment. > > I *have* been able to get the displacement all the way down eventually, > though it gets harder the larger the system and the greater the constraints. > > Are you constraining cartesian co-ordinates? I suggest trying to constrain > a minimal number of internal co-ordinates for key heavy atoms. > > Hope this helps a little, > > Jeremy > ---------------------------------------------------------------------- > Jeremy Greenwood jeremy.greenwood@i.am > Department of Medicinal Chemistry bh +45 35306117 > Royal Danish School of Pharmacy fx +45 35306040 > Universitetsparken 2, DK-2100 Copenhagen, Denmark ah +45 32598030 > ---------------------------------------------------------------------- > > > == Response #2: Nicolas Ferre ================ > Hi Gerald, > > I'm very surprised by the output you give. Normally the total Delta X is > the sum of the 2 previous one, Linear and Quad. And when you freeze some > variables, all of theses 3 quantities have to be zero. I've performed > some calculations on crambin, allowing only 3 residues to move, so > freezing other 43 residues (it's a QM/MM calculation) and the output > looked fine. > In your case, it seems to me that only R1 to R5 are really frozen, but > that atoms 1, 11, 38, 48 and 50 are translated by the vector (2.501, > -8.90656,8.10547) in bohrs. I don't understand why. > Are you sure that these atoms don't move ? Did you check the values > during optimization ? > How did you freeze the coordinates of these atoms ? With a "-1" in the > How did you freeze the coordinates of these atoms ? With a "-1" in the > Z-Matrix or with the MODREDUNDANT keyword ? Did you use the NOSYM keyword > (optimization in the input orientation or in the standard orientation) ? > > Nicolas > > > == Response #3: Stefan Fau ================================= > Hi, > > an interesting practical problem... Here are my 2 cents, > comments invited. > > to 2) and 3) > Delta X and New X are the differences and new values > of the geometry descriptors. G98 allows mixed geometry > specifications using internal coordinates together with > cartesian coordinates. Freezing cartesian coordinates is > equivalent to freezing all internal coordinates between this > group of atoms (provided you have only one spatial group > of frozen atoms). > > If you have two clusters of frozen atoms linked by a > unconstrained part of the molecule, one way of optimizing > the molecule is to allow movement of the two clusters with > respect to each other. I don't think G98 does it this way, but > you can check by tracking the distances between atoms of > different frozen clusters over the optimization. > > Another reason for blockwise movement of the frozen atom > clusters is that sometimes the standard orientation of the > molecule flips from one optimization step to the next. (I think > that the standard orientation is determined by the moments of > inertia. Sometimes two of them are very similar and change > sequence from one geometry to the next.) > > I think the list of geometry variables contains both the original > set of geometry descriptors and the redundant internal coordinates. > I am not sure if original descriptors that are NOT redundant internal > coordinates enter in the computation of the displacement used as a > convergence criterion. You may want to browse the CCL archives: > I faintly remember a mail from Gaussian (maybe Doug Fox) on that > topic a year or two ago. > > to 1) > Vibrational frequencies are molecular properties. They all depend > on all of the molecule. At the same time, some are quite specific > for functional groups. Therefore the influence of other parts of the > molecule must be small, probably rapidly declining with distance. > > The common implementation of harmonic frequencies requires the > forces (=first derivatives of the energy w.r.t. nuclear positions) to > be > (nearly) zero. I don't know how the second derivatives change with > displacements that cause nearly no forces, but the displacements must > be along floppy modes with very small force constants and frequencies. > I suppose that big residual displacements will mainly influence small > frequencies. > > In the harmonic approximation you may diagonalize the matrix of > mass weighted second derivatives of the energy with respect to > nuclear coordinates and get the vibrational modes and vibrational > energies. Like with the solution of the HF equation, you can apply > unitary transformations to get a localized description (ideally only > one internal coordinate varying in a mode) that is not diagonal > anymore. The off-diagonal matrix elements describe how the localized > modes influence each other. > > More to your problem: I think you can use frequencies that have NO > movement of frozen atoms in the corresponding vibrations. You may > try to block-diagonalize the matrix with a unitary transformation that > constricts vibrations with motion of frozen atoms in one block. Since > the coupling elements with the other block would be zero, the > frequencies > >from that block would be valid. > > One caveat: > When you truncate a molecule, you omit interactions between the cut > off > parts as well as interactions between the cut off and the retained > part of > the molecule. If you have to freeze clusters of atoms to force the > truncated > molecule into the original configuration, some of the omitted > interactions > were important. I would like to make sure that none of the cut off > parts had > a significant interaction with the interesting part of the molecule. > > I hope it answers some of your questions, > Stefan > ______________________________________________________________________ > Dr. Stefan Fau | fau@qtp.ufl.edu > Quantum Theory Project | (352) 392-6714 > University of Florida > Gainesville, FL 32611-8435 > > =============================================================================== > > > > -= This is automatically added to each message by mailing script =- > CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins > MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH > CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 > Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net > > > > > ---------------------------------------------------------------------- Gerald H. Lushington Ph: 614-292-6036 Research Specialist Fax: 614-292-7168 OSC / PET-CCM e-mail: gerald@ccl.net 1224 Kinnear Road http://www.arl.hpc.mil/PET/cta/ccm Columbus OH 43212-1163 http://www.asc.hpc.mil/PET/CCM From chemistry-request@server.ccl.net Thu Dec 7 16:00:29 2000 Received: from server.biokemi.su.se (server.biokemi.su.se [130.237.179.5]) by server.ccl.net (8.8.7/8.8.7) with SMTP id QAA09680 for ; Thu, 7 Dec 2000 16:00:28 -0500 Received: from mail.biokemi.su.se (unverified [130.237.179.175]) by server.biokemi.su.se (EMWAC SMTPRS 0.83) with SMTP id ; Thu, 07 Dec 2000 22:00:46 +0100 Sender: zhang@ccl.net Message-ID: <3A2FFA4F.C22B6E90@mail.biokemi.su.se> Date: Thu, 07 Dec 2000 21:59:59 +0100 From: Xiao-Ping Zhang Reply-To: zhang@biokemi.su.se X-Mailer: Mozilla 4.72 [en] (X11; U; Linux 2.2.14-12 i686) X-Accept-Language: en MIME-Version: 1.0 To: chemistry@ccl.net Subject: install prng on IBM SP2 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Dear CCLers, I would like to install ECEPPAK on IBM SP2. I should install prng (Parallel Random Number Generators, written by Prof. Mal Kalos) first. According to the description at the end of ECEPPAK manual: , there is a Makefile.IBMSP2 for this system somewhere. Unfortunately, I did not find this file. Therefore I was not be able to install PRNG and ECEPPAK. If someone knows where this file is or has experience on installation of ECEPPAK on IBM SP2, I would be grateful to get your help. Thank you advance for your help. Sincerely, Xiao-Ping Zhang ****************************** Xiao-Ping Zhang Department of Biochemistry Arrhenius Laboratories of Natural Sciences Stockholm Universities 106 91 Stockholm Sweden Phone: 046-08-162472 /162582 Fax: 046-08-153679 e-mail: zhang@biokemi.su.se