From chemistry-request@ccl.net Sat Aug 7 07:29:19 2004 Received: from aiolos.otenet.gr (aiolos.otenet.gr [195.170.0.23]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id i77CTHHX026629 for ; Sat, 7 Aug 2004 07:29:18 -0500 Received: from home (athe530-b239.otenet.gr [212.205.219.239]) by aiolos.otenet.gr (8.12.11/8.12.11/Debian-3) with SMTP id i77Cao0F031706 for ; Sat, 7 Aug 2004 15:36:53 +0300 Message-ID: <001401c47c7b$3acea7a0$efdbcdd4@home> From: "Kostas Papavasileiou" To: Subject: Help in Gaussian 98 Date: Sat, 7 Aug 2004 15:36:54 +0300 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-7" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.3790.181 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.3790.181 X-Spam-Status: No, hits=5.1 required=7.5 tests=FROM_ENDS_IN_NUMS, FROM_HAS_ULINE_NUMS,RCVD_IN_DYNABLOCK,RCVD_IN_SORBS autolearn=no version=2.61 X-Spam-Checker-Version: SpamAssassin 2.61 (1.212.2.1-2003-12-09-exp) on servernd.ccl.net Dear All, I tried to do the following calulation: CIS(full)/cc-pVTZ Xanthene geometry optimization and frequencies calculation of 1st excited state. in Gaussian98 and received the following error message: Inverted reduced A of dimension 886 with in-core refinement. Max Tx.T = 0.000249024916 for Degree = 3 Tx not orthogonal to T. Error termination via Lnk1e in What is wrong here? Why is this appearing Any help will be very appreciated. Thank you in advance!! Kostas Papavasileiou University of Ioannina Greece From chemistry-request@ccl.net Sat Aug 7 03:02:58 2004 Received: from collapsed.wormhole.hu (collapsed.wormhole.hu [195.70.35.130]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id i7782uHX024921 for ; Sat, 7 Aug 2004 03:02:56 -0500 Received: from zeus.canet.hu ([212.97.0.169] helo=Moon) by collapsed.wormhole.hu with asmtp (Exim 4.34 #1 (Debian)) id 1BtMHg-0003Md-Fq; Sat, 07 Aug 2004 10:10:32 +0200 Received: from szilva (helo=localhost) by Moon with local-esmtp (Exim 3.36 #1 (Debian)) id 1BtMDm-00015i-00; Sat, 07 Aug 2004 10:06:30 +0200 Date: Sat, 7 Aug 2004 10:06:30 +0200 (CEST) From: Szilveszter JUHOS X-X-Sender: szilva@Moon To: "Pandey, Jaya" cc: "'chemistry*at*ccl.net'" Subject: Re: CCL:Binding energy In-Reply-To: <73564B8DE398D61180500008C75D7E49033E6091@NTMAIL03> Message-ID: References: <73564B8DE398D61180500008C75D7E49033E6091@NTMAIL03> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: Szilveszter Juhos X-Authenticated-Sender: 2cea3f6ec7de544d95b27aa98c46f37a X-Spam-Status: No, hits=-0.5 required=7.5 tests=SMILEY autolearn=no version=2.61 X-Spam-Checker-Version: SpamAssassin 2.61 (1.212.2.1-2003-12-09-exp) on servernd.ccl.net On Fri, 6 Aug 2004, Pandey, Jaya wrote: > I screened about 1900 small ligands using AutoDock and the lowest energy of > binding I could get was -10.00. Is it worth trying this ligand in the lab? Dear Jaya, Welcome to the docking club! ;) As you are writing -10, it does not mean too much (like I am saying that is is 22 C outside, but is it cold or warm? Depends... ) . My (personal) view would be that i) 1900 candidates are rather few if you are not working with a ligand set that is focussed on your particular target. For example if it a carefully selected kinase library and you are targeting a kinase, probably this amount is a minimum to start docking with. I would consider 100K filtered ligand at least or more. ii) lowest energy does not necessarily mean it is the best binding ligand: the relationship between the calculated and real dG is not always linear and there is a quite a lot error both in measuring and calculating dG iii) the ligand can be too "sticky": full of H-bond donors/acceptors and happy to bind anywhere to polar surface with high affinity (pre-filtering comes here) iv) check carefully wether are there "bad contats" not penalised by the scoring function. Typical things are lone pairs facing to each other, voids between the ligand and the target v) the ligand can face its apolar part towards the solvent vi) if there is a chance do not try only the best scoring one but a diverse subset of the good scoring ones. and it goes on like this. Sometimes it is worth to have a look to known complexes (like the GOLD/Astex and the Glide set), check the polarity of the cavity, position of aromatic rings/stacks etc. just to train you eyes. > Another query is that how to improve the binding energy. What about having few candidates, creating a diverse set based on these molecules and re-docking them? > Please advice and help, what kind of energy one should look for to get a > lead? This a multi-million dollar question :) .. I wish I knew the answer! I would say (again: personel view) docking is not the best for lead development but god for hit-identification. Once you have a dozen or two diverse hits you can play a lot with other tools/experiments to get two-three leads. Szilva From chemistry-request@ccl.net Sat Aug 7 07:35:50 2004 Received: from aiolos.otenet.gr (aiolos.otenet.gr [195.170.0.23]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id i77CZmHX027105 for ; Sat, 7 Aug 2004 07:35:48 -0500 Received: from home (athe530-b239.otenet.gr [212.205.219.239]) by aiolos.otenet.gr (8.12.11/8.12.11/Debian-3) with SMTP id i77ChOEE000931 for ; Sat, 7 Aug 2004 15:43:25 +0300 Message-ID: <002101c47c7c$24017240$efdbcdd4@home> From: "Kostas Papavasileiou" To: Subject: Help in Gaussian 98 Date: Sat, 7 Aug 2004 15:43:28 +0300 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-7" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.3790.181 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.3790.181 X-Spam-Status: No, hits=5.1 required=7.5 tests=FROM_ENDS_IN_NUMS, FROM_HAS_ULINE_NUMS,RCVD_IN_DYNABLOCK,RCVD_IN_SORBS autolearn=no version=2.61 X-Spam-Checker-Version: SpamAssassin 2.61 (1.212.2.1-2003-12-09-exp) on servernd.ccl.net Dear All, I tried to do the following calulation: CIS(full)/cc-pVTZ Xanthene geometry optimization and frequencies calculation of 1st excited state. in Gaussian98 and received the following error message: Inverted reduced A of dimension 886 with in-core refinement. Max Tx.T = 0.000249024916 for Degree = 3 Tx not orthogonal to T. Error termination via Lnk1e in What is wrong here? Why is this appearing Any help will be very appreciated. Thank you in advance!! Kostas Papavasileiou University of Ioannina Greece > >