From chemistry-request@server.ccl.net Fri Jun 11 04:15:56 1999 Received: from gateway.bioreason.com (bioreason.com [207.108.245.3]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id EAA13772 for ; Fri, 11 Jun 1999 04:15:56 -0400 Received: from bioreason.com (val [192.168.1.51]) by gateway.bioreason.com (8.9.3/8.9.3) with ESMTP id CAA18772 for ; Fri, 11 Jun 1999 02:01:20 -0600 Sender: dalke@bioreason.com Message-ID: <3760C3CA.C4475CD8@bioreason.com> Date: Fri, 11 Jun 1999 02:07:38 -0600 From: Andrew Dalke Organization: Bioreason, Inc. X-Mailer: Mozilla 4.05C-SGI [en] (X11; U; IRIX64 6.5 IP30) MIME-Version: 1.0 To: chemistry@ccl.net Subject: maximum common substructure Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hello, Some time back I asked about available software or methods to find the 2D maximum common substructure (MCS) between a set of compounds. Here are the belated results of that query. I started off looking for clique based solutions, as suggested in http://baker.mcs.kent.edu/paper/ECCC5.html. It turns out that those solutions are not very useful for cyclical structures because the correspondence graph becomes very dense, and hence very hard to use. (BTW, if anyone is interested in clique detection codes, the Bron-Kerbosch is available on-line from netlib at http://www.netlib.org/tomspdf/457.pdf and Ram Samudrala has converted the code into C as part of RAMBIN, at http://www.twisted-helices.com/computing/rambin/rambin.html . There are also some newer public domain algorithms available at ftp://dimacs.rutgers.edu/pub/challenge/graph/solvers/ which are somewhat faster than Bron-Kerbosch for the graphs we were testing.) I received several relevent responses from my CCL query. Yvonne Martin pointed out that DISCO uses a clique based solution pharmacophore mapping in multiple molecules, and mentioned that clique solutions are not used for 2D MCSes. Wally Reiher suggested contacting Peter Willett about the subject because of his previous work in the matter. David Wild also corrected my misunderstanding that clique codes were used for 2D MCS searches. He pointed out some other people who may have worked on the multiple-compound MCS problem, saying: > Peter Johnson's group in Leeds may have done some kind of > muliple-MCS that was never published, and perhaps someone in Johann > Gasteiger's group too, although I don't think there were any > publications from that either. Finally, David Lowis pointed out that > Tripos has a piece of code (mcs.exe) which given a set of molecules > will determine the maximum common substructure (MCS) amongst them. However, it turns out that is part of the HQSAR package for which we don't have a license and which costs too much for our needs. From this response I started digging some more. I would like to point out a very complete bibliography of subgraph isomorphism papaers exists at http://www.ics.uci.edu/~eppstein/bibs/subiso.bib . By searching that list and with private email to Peter Willett I only found one paper which seems applicable to the 2D multiple compound MCS problem, which is: @article{BaySimJoh-CICS-92, title = {{An algorithm for the multiple common subgraph problem}}, author = {D. M. Bayada and R. W. Simpson and A. P. Johnson and C. Laurenco}, journal = {J. Chemical Information and Computer Sciences}, volume = {32}, pages = {680--685}, year = {1992}, abstract = {The multiple largest common subgraph (MLCS) problem is presented, ....} } That paper describes both a pairwise MCS algorithm and the extension of the pairwise solution to a set of compounds. While waiting to get that paper we started a pairwise implementation using a graph search with backtracking implementation based on the general ideas described in: Backtrack Search Algorithms and the Maximal Common Subgraph Problem James J. McGregor, Software-Practice and Experience, vol. 12, 23-34 (1982). For those interested, a description of our algorithm and a reference implementation in Python is available from http://starship.python.net/crew/dalke/MCS/ Some of the compounds used in the testing of the code are proprietary in nature so parts of the test code and documentation have been removed. My apologies for a possibly more confusing distribution. The general form of our method is similar to that given in the Johnson paper, since their algorithm also comes from McGregor. However, I believe we provide a better description of the algorithm details and our code can be used to generate all unique maximal mappings directly; which isn't apparent from the Johnson paper description. On the other hand, they describe some optimization techniques which we don't (yet) use. Andrew Dalke dalke@bioreason.com From chemistry-request@server.ccl.net Fri Jun 11 11:08:21 1999 Received: from theo1.theochem.tu-muenchen.de (theo2.theochem.tu-muenchen.de [129.187.191.1]) by server.ccl.net (8.8.7/8.8.7) with SMTP id LAA17857 for ; Fri, 11 Jun 1999 11:08:18 -0400 Received: from theo350.theochem.tu-muenchen.de (theo350.theochem.tu-muenchen.de [129.187.157.44]) by theo1.theochem.tu-muenchen.de (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id RAA18175 for <@theo1.theochem.tu-muenchen.de:chemistry@ccl.net>; Fri, 11 Jun 1999 17:01:29 +0200 Received: (from gisdakis@localhost) by theo350.theochem.tu-muenchen.de (950413.SGI.8.6.12/950213.SGI.AUTOCF) id RAA03310 for chemistry@ccl.net; Fri, 11 Jun 1999 17:01:24 +0200 From: "Philip Gisdakis" Message-Id: <9906111701.ZM3308@theo350.theochem.tu-muenchen.de> Date: Fri, 11 Jun 1999 17:01:24 -0600 X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: chemistry@ccl.net Subject: Orbital Viewer Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Dear Collegues, we are seeking for are free orbital viewer for Windows NT, which can display MOs calculated by Gaussian. We need it for educational purposes. Also suitable would be a program, which can build a VRML file, as molden does, but in command line mode. We would run that program on our Unix Server and the VRML-file is just transferred to the students PC, which will display it. Thanks in advance Philip -- Philip Gisdakis o-----------------------------------------------------------------o / Philip Gisdakis gisdakis@theochem.tu-muenchen.de / \ / Theoretical Chemistry Phone: +49 +89 289 13618 (Univ.) / o /Technische Universitaet Muenchen 67908232 (Privat) / / / Lichtenbergstrasse 4 / / / D - 85747 Garching / / o-----------------------------------------------------------------o / \ \ / o-----------------------------------------------------------------o From chemistry-request@server.ccl.net Wed Jun 9 11:00:10 1999 Received: from hotmail.com (f238.hotmail.com [207.82.251.129]) by server.ccl.net (8.8.7/8.8.7) with SMTP id KAA22332 for ; Wed, 9 Jun 1999 10:59:59 -0400 Received: (qmail 16417 invoked by uid 0); 9 Jun 1999 14:51:18 -0000 Message-ID: <19990609145118.16416.qmail@hotmail.com> Received: from 134.129.184.39 by wy1lg.hotmail.com with HTTP; Wed, 09 Jun 1999 07:51:18 PDT X-Originating-IP: [134.129.184.39] From: Rajeev Pandey To: chemistry@ccl.net Subject: cubic and quartic force constants Date: Wed, 09 Jun 1999 07:51:18 PDT Mime-Version: 1.0 Content-Type: text/plain; format=flowed Dear CCLers, I am looking for references, for calculating cubic and quartic force constants numerically from analytic second derivatives. I would appreciate if someone has any references. Thanks in advance. Rajeev. ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com From chemistry-request@server.ccl.net Wed Jun 9 19:04:10 1999 Received: from relay2.scripps.edu (relay2.scripps.edu [137.131.200.30]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id TAA26986 for ; Wed, 9 Jun 1999 19:04:10 -0400 Received: from elvis.scripps.edu (elvis.scripps.edu [137.131.130.25]) by relay2.scripps.edu (8.9.3/TSRI-3.0.2r) with ESMTP id PAA18792 for ; Wed, 9 Jun 1999 15:57:44 -0700 (PDT) Received: from scripps.edu (localhost [127.0.0.1]) by elvis.scripps.edu (8.9.2/TSRI-3.0) with ESMTP id PAA100080 for ; Wed, 9 Jun 1999 15:57:44 -0700 (PDT) Sender: yadavm@scripps.edu Message-ID: <375EF168.29495590@scripps.edu> Date: Wed, 09 Jun 1999 15:57:44 -0700 From: Maneesh Yadav X-Mailer: Mozilla 4.07C-SGI [en] (X11; I; IRIX64 6.5 IP30) MIME-Version: 1.0 To: chemistry@server.ccl.net Subject: Some questions about Dock 4.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hello all, I am new to the list, so if this discussion has recently come up, a pointer would be appreciated. I am comparatively new to this field, so the questions I have may be classified as beginner status, but please bear with me. We are doing some experiments of substrate binding to internal protein cavities, and for a screening of possible substrates we are using Dock 4.0. I have go the program working for single ligand binding (and has given me some fair results) but I do have some questions, any advice on the following would be appreciated. A) Multiple Ligands I have saved an FDAT file from the CSD, and converted to PDB format (via fdat2pdb). Now, when I convert to mol2 via pdb2syb, I get messages like: pdb2syb binder.pdb H4 in . for C- 2 NI1 in . for C- 2 O5 in . for C- 2 O6 in . for C- 2 O7 in . for C- 2 5 isolated atom(s) omitted from . for C- Error--no bond length information for atom A in . for C- Skipping . for C- C3 in . for C- 3 C4 in . for C- 3 C5 in . for C- 3 3 isolated atom(s) omitted from . for C- I am assuming this is due to the fact that the respective crystal structures are somehow incomplete? And then when I try to pass these to Dock, it gives errors like, WARNG read_mol2: Improper ATOM record substructure id for _ . for C- in binder .mol2 and no real output. Is this something unpreventable or have their been some changes in the database format of CSD? Another thing is that I am currently querying the CSD for structures that have between 1-25 C atoms, what I really want is to get a large number and variety of small molecules that could fit in a sphere of radius ~4.5A (about the cavity radius). Does anyone have a better method of a volume constraint based query (even for another database)? B) Chemical Screening/Matching When I insert the color table into the the .sph file at the appropriate place, color hydrophobic 1 (each sphere in my cluster is labeled hydrophobic) and then label the last (color) column correspondingly 1; I turn chemical matching on and Dock says that the "label hydrophobic is undefined" and doesn't read the file BTW I am pointing Dock to the right definition tables. Chemical scoring at least reads the file, but I am suspicious of the results as a hydrophobic molecule generated a positive chemical score and a charged molecule has given me a sizable negative score! Am I doing something obviously wrong? I have also tried directing my questions to members of the Kuntz group, but they seem rather busy right now and have been unable to respond. Many Thanks in Advance, Maneesh Yadav yadavm@scripps.edu From chemistry-request@server.ccl.net Thu Jun 10 16:26:05 1999 Received: from helios.qc.ag-berlin.mpg.de (helios.QC.AG-Berlin.MPG.DE [141.20.74.13]) by server.ccl.net (8.8.7/8.8.7) with SMTP id QAA07975 for ; Thu, 10 Jun 1999 16:26:04 -0400 From: jrh@qc.ag-berlin.mpg.de Received: from helios.QC.AG-Berlin.MPG.DE by helios.qc.ag-berlin.mpg.de (AIX 3.2/UCB 5.64/4.03) id AA20794; Thu, 10 Jun 1999 22:19:26 +0200 Date: Thu, 10 Jun 1999 22:19:26 +0200 Message-Id: <9906102019.AA20794@helios.qc.ag-berlin.mpg.de> To: CHEMISTRY@ccl.net Subject: Viewmol 2.2.1 has been released. What is Viewmol ? Viewmol is a graphical front end for computational chemistry programs. It is able to graphically aid in the generation of molecular structures for computations and to visualize their results. The program's capabilities include: - Building and editing of molecules - Visualization of the geometry of a molecule - Tracing of a geometry optimization or a MD trajectory - Animation of normal vibrations or to show them as arrows - Drawing of IR, Raman, and inelastic neutron scattering spectra - Drawing of an MO energy level or density of states diagram - Drawing of basis functions, molecular orbitals, and electron densities - Display of forces acting on each atom in a certain configuration - Drawings generated by Viewmol can be saved as TIFF, HPGL, or PostScript files - Animations of normal modes can be converted to a video file (MPEG), e. g. for inclusion into World Wide Web documents (requires additional programs available on the Internet) - Interface to the freeware ray tracing program Rayshade (input file generation and use of Rayshade from within Viewmol) - Input and output in a variety of formats, new formats can be added easily by the user At present Viewmol includes input filters for Discover, DMol/DSolid/DMol3, Gaussian 9x, Gulp, Mopac, and Turbomole outputs as well as for PDB files. Structures can be saved as MSI car-files, MDL files, and Turbomole coordinate files. Viewmol's file format has been added to Babel so that Babel can serve as an input as well as an output filter for coordinates making all formats Babel can handle accessible by Viewmol. What is new in version 2.2.1 ? - All bugs discovered in version 2.2 have been fixed - Van der Waals radii (and therefore connectivity) can now be changed interactively - Optimization histories/MD trajectories can now be animated - improved PDB input filter (courtesy of Andrew Dalke) - precompiled binaries for OSF1 are back (courtesy of Pablo Vitoria Garcia) - Viewmol now compiles on Suns (courtesy of Keith Refson) - Bindings and icons for KDE included (KDE will recognize all file types supported by Viewmol, display special icons, and start Viewmol on clicking on the icon) - The Linux version of Viewmol is now also available as RPM What operating systems does Viewmol run on ? Viewmol was developed using Linux, but it was also ported (or better recompiled) on a number of other operating systems. Currently, pre- compiled binaries are available for Linux (dynamically linked with Lesstif, statically linked with Motif), AIX, IRIX, and OSF1. The program should also run on HP-UX, FreeBSD, and SunOS but due to a lack of direct access to such hardware I couldn't test it. Where do I get Viewmol ? Viewmol can be downloaded from the Computational Chemistry List archive (USA): - http://server.ccl.net/chemistry/resources/software/SOURCES/C/viewmol/ The source code and the precompiled binaries for Linux will also be available from Sunsite (USA) or your friendly neighbourhood mirror: - file://sunsite.unc.edu/pub/Linux/apps/graphics/visualization For the first two sites it may take a few days before the site administrators have moved the code to the given location. Jörg-Rüdiger Hill (jrh@qc.ag-berlin.mpg.de) From chemistry-request@server.ccl.net Fri Jun 11 04:15:58 1999 Received: from gateway.bioreason.com (bioreason.com [207.108.245.3]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id EAA13777 for ; Fri, 11 Jun 1999 04:15:57 -0400 Received: from bioreason.com (val [192.168.1.51]) by gateway.bioreason.com (8.9.3/8.9.3) with ESMTP id CAA18776 for ; Fri, 11 Jun 1999 02:01:22 -0600 Sender: dalke@bioreason.com Message-ID: <3760C3CC.9FF7A619@bioreason.com> Date: Fri, 11 Jun 1999 02:07:40 -0600 From: Andrew Dalke Organization: Bioreason, Inc. X-Mailer: Mozilla 4.05C-SGI [en] (X11; U; IRIX64 6.5 IP30) MIME-Version: 1.0 To: chemistry@ccl.net Subject: Announcing PyDaylight 0.5 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit PyDaylight 0.5 Bioreason , a leading provider of chemoinformatics knowledge discovery services. has developed an interface between the Daylight toolkit and the Python programming language and is making this package freely available under the LGPL license. Additional information is available from: http://starship.python.net/crew/dalke/PyDaylight/ and from my MUG'99 presentation at: http://www.daylight.com/meetings/mug99/ The end result is that new chemical analysis algorithms can be developed in a much shorter time and with fewer programming errors. Python is an interpreted, interactive, object-oriented programming language with modules, classes, exceptions, very high level dynamic data types, and dynamic typing. It is a superb tool for rapid application prototyping and development. For more information, see http://www.python.org/ . The Daylight toolkit is sold by Daylight Information Systems (see http://www.daylight.com/). The toolkit includes C/FORTRAN libraries for manipulating molecular topologies, depicting the results, conducting subgraph searches, and more. (We don't get any money from them for promoting their software :) The PyDaylight interface builds on Roger E. Critchlow Jr.'s DaySWIG package but adds a true object-oriented interface layer above the underlying C libraries as well as iterators, automatic garbage collection, and exception handling. We develop all of our applications using PyDaylight and it has been very stable and useful. Since it has been an in-house project there is very little end-user documentation or tutorials, which is why it is labeled 0.5 instead of 1.0. Thus, this release shouldn't be used by someone who doesn't like reading source code to figure out what's going on. However, I would like to know who might be interested in introductory material, so if you plan on using the Daylight toolkit and would like to try out the PyDaylight interface, please drop me a message so I can better gauge what to work on next. Andrew Dalke dalke@bioreason.com From chemistry-request@server.ccl.net Fri Jun 11 04:36:43 1999 Received: from rs5.thch.uni-bonn.de (rs5.thch.uni-bonn.de [131.220.44.15]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id EAA14053 for ; Fri, 11 Jun 1999 04:36:39 -0400 Received: (from bernd@localhost) by rs5.thch.uni-bonn.de (8.8.8/8.8.5) id KAA17242 for chemistry@www.ccl.net; Fri, 11 Jun 1999 10:29:41 +0200 From: Bernd Engels Message-Id: <199906110829.KAA17242@rs5.thch.uni-bonn.de> Subject: summary_errors in comp chem. To: chemistry@server.ccl.net (Computational Chemistry List) Date: Fri, 11 Jun 1999 10:29:40 +0200 (MES) X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Below please find a summary of the responses I got to the following question: >I am looking for systems which are theoretically >difficult. I am interested in all kind of systems, >i.e. in systems which makes problems for DFT and/or >Coupled Clusters and/or Moller-Plesset etc. >there should be a lot of systems which are >really problematic and (what a pity) where the >problems where never published. Thanks a lot to all who have answered. It was really interesting. Bernd Engels ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Dear Prof. Engels, I have started to put together a web accessible database of molecules and particular theoretical methods which produce poor results. It doesn't have many species in it yet. I have it running on a small web server which is currently at: http://h194161.nist.gov/sicklist/sicklist.html I hope to move it to a better server in the next month or so at which point he address will change. It is set up so that anyone can make a contribution. If the number of entries gets large enough I can add methods for searching by molecule or method or whatever. It may have some suggestions of the sort you are looking for. I was hoping to catch problems that don't get published. I would welcome any comments you have about the Sicklist database. Please do not give out the address. When I get it moved to the better web server I will make a posting on the CCL. Russ Johnson ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ The base properties of N2O are a problem. See J. E. Del Bene, E. Stahlberg, and I. Shavitt Int. J. Quantum Chem. Quantum Chem. Symp. 24, 455 (1990). This paper has references to some earlier work as well. Regards, Janet Del Bene ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ hi Bernd, You can try to calculate FeO - both ground and excited state. good luck! Pradipta ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Bernd, the trichlorine radical, Cl3, seems to be quite difficult; see Kaledin et al., J. Chem. Phys. 108, 2771 (1997). Best regards, Ole Swang Status: RO ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Dear Bernd, At the last ACS meeting, Karl Irikura reported that his group at NIST is developing several resources to be made readily available in support of QC applications: http://www.nist.gov/compchem One of these developments is to collect examples of known failures of specific methods into a "Quantum Chemistry Sicklist Database." This sounds very much like what you are looking for: >I am looking for systems which are theoretically >difficult. I am interested in all kind of systems, >i.e. in systems which makes problems for DFT and/or >Coupled Clusters and/or Moller-Plesset etc. >there should be a lot of systems which are >really problematic and (what a pity) where the >problems where never published. I think part of their interest is just what you described: that it is a pity they were never published so researchers can avoid pitfalls and theoreticians can focus on where methods need improvement. You should get in touch with Karl's group at NIST, and see what they have collected so far. EC --- Ernest Chamot Chamot Laboratories, Inc. 530 E. Hillside Rd. Naperville, IL 60540 (630) 637-1559 echamot@chamotlabs.com http://www.chamotlabs.com/cl ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Bernd, Here are a few suggestions. Transition metals, particularly open shell and multiple bonded ones. Au which is sensitive to relativistic effects. F2 and FOOF which are correlation sensitive. Jahn-Teller distortions which are correlation sensitive. Please post a summary to the CCL. Dave Young youngd2@mail.auburn.edu From chemistry-request@server.ccl.net Fri Jun 11 06:23:26 1999 Received: from rtec01.acadia-pharm.com ([194.255.124.226]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id GAA14838 for ; Fri, 11 Jun 1999 06:23:24 -0400 Received: from thc.acadia-pharm.com (THC [194.192.178.248]) by rtec01.acadia-pharm.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2232.9) id M1ABGNPY; Fri, 11 Jun 1999 12:16:53 +0200 Date: Fri, 11 Jun 1999 12:14:05 +0200 (CEST) From: Jan Pedersen To: chemistry@ccl.net Subject: CCL:maximum common substructure - the code Message-ID: MIME-Version: 1.0 Content-Type: MULTIPART/MIXED; BOUNDARY="-1060964174-1454727082-929096045=:16828" This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. Send mail to mime@docserver.cac.washington.edu for more info. ---1060964174-1454727082-929096045=:16828 Content-Type: TEXT/PLAIN; charset=US-ASCII Greetings, Please find attached Graph.tar containing the necessary code for maximum common sub-structure identification using Bron & Kerbosh. Here is the Bron & Kerbosh code to do the job - we have used Ram's code which is originally an implementation by Jean Francois Gibrat - then at the NCBI. The code has been slightly modified and you have to be aware of it's interpretation of a common sub-graph, that is: A maximally connected subgraph does not imply a connected substructure. Second, you can only compare molecules where all atoms are connected in a single graph, that is you have to be careful with the treatment of salts. Anyhow, that aside it is really fast and we use it routinely to evaluate large libraries of structures against our own in-house screening library (currently 160,000 compounds). Unfortunately I am unable to send the complete implementation at this moment, but nimble (C-)fingers will make the following code work in a jiffy. The procedure is: 1) The routine BestFitSmall() compares two structures and returns the largest Clique, that is the maximum common sub-graph. 2) The routine MakeGraph() makes the graphs for each of the two structures to be compared. Once that is done a DockingGraph() can be made for comparison. 3) FindCliques() then runs Extend() in a recursive branch and bound fashion to find all the clicques. 4) The maximum common sub-graph is stored in external variables declared in the include file GraphStatistics.h All of this code comes ofcourse without any form of warranty and ``as-is'' we will ofcourse be interested in any extensions or interesting science you manage to do with it. The code comes out of my own molecular simulation program 'J' that is why the code may seem somwhat torn-out. I have just tar-ed down everything in my ./Graph source directory, and ./Types directory. If you need any further routines please let me know. I hope I will be able to mail out a free standing version of the program within the next month. Regards and good luck -- Jan ----------------------------------------------------------- Dr. Jan T. Pedersen Head, Computational Chemistry and Chemical Informatics Acadia Pharmaceuticals Fabriksparken 58 2600 Glostrup, DENMARK email : jan@Acadia-Pharm.com phone : +45 43 29 30 05 FAX : +45 43 29 30 30 Lore : http://www.acadia-pharm.com ----------------------------------------------------------- ---1060964174-1454727082-929096045=:16828 Content-Type: APPLICATION/x-gunzip; name="Graph.tar.gz" Content-Transfer-Encoding: BASE64 Content-ID: Content-Description: Content-Disposition: attachment; filename="Graph.tar.gz" H4sICKXWYDcAA0dyYXBoLnRhcgDtPWtX20iy+9X+FZ3MYzExxpJtnOAw9xBi 8lheB8jM7s3lcGRbBoEteSUZwuzkv9+q6oe6JRkbEkhmVj0ZbPejurq7uqr6 UdVvQmdyvro5Gh2dB2HsRvGBE59Htf7fvmKw6vW1ZpP9jTHWXuOflviNwW5b FmNr7UbdatlrDRtSG22r/jdW/5pIzArTKHZCxv524fhz8rlh9BgIPW5YXf6a oVxmqfD3939nK+xfbsw2/SA+d0O2G4zc/nQEnb7r9s8d3+tH7CAMzkJnnC2u haV+hVkvXrTYe8dnx+zAHcCAuH65XD4+9yI24SAYfIWKmOczqI1Npr2R12eD YOxAjOMP2Ni5oQw9l8GQDlgQsoEXxaHXm8buoNy7gVw3ge+yay8+D6YxgbkO vTh2fTZxw7EXRV7gs2BIKc4UcoU1dgzfh1O/H0OaM/Lim7LIcH0O7cVaJk4Y R7JYCtuzqRM6fuy6g1q5/FVHZLnMfnzrOtBXP8K3Q/fKQ+x//LqVrJbLP3h+ fzQduOzpG+QotfOn5fJV4A1YmrUs9QM/ihnlYsuTM/ysyl+hO5j23UGl/J+y 58fM94OBG1W96kX1skMxg1PvooO1DQfukL1+9aY8nISQMGRLUTyA4ao+hUj2 yy/Zeiv/5z+tdMo/uP7AG5Y56I0lgUGlxiM65SEM1ZK3Ue94L0WU9+xZBQiz 9B+izhJluIAMFzLDBc8AaSIPfvOGCGbjoqJoulSS7Vv5Zfejt8xLP/NOAJYq 5o4il2FZjpeR8+Lkl/pccBcnG5YJbpESdQwd0YjP2NrPZd4Vl4DdpWzpZU5X 5PWV2RWix7xnVk6fYQaVF3/gEG/kInp58kyPv0wa0NHKY+chjJczGqv3oFYx /ZzVP5zsZma90MbSzFrivWl8VR18LyoGbjpA9vTEIOfP5Xvy/zdS/gd9PnG/ quTnYZ78bzXXNPnfhlS72W4U8v8xQiH/C/n/xWGG/KdvyyxhLkuJUEcRb+gA G3sfdna4kPeqvo8ycUEGKYGbAr6MckBI0Y0ljkpl7GD2pcj73Q2GPLJSqWxQ 3ZUySINSNwyPQ6fvsqWnCeinVbb/4Xh/e7e7CxWUQjeehj5bolLw+3NZCmuh VUhdwvfZBlsSssGqLCdfOwZ+IEM2qG+WK32OIaF4GlegRKUq0IUMKWQznQLY A6WtG70C0++D7/SACOOAEXQndtnYHQfhDe+xRRotZNgwdF2BdH5PGCpBog5g OHLjvYBLsZ919eakUrn7WAsygrb+NBAiERpTFcTVWQyG6FjRHgRVNeMAP4pG 0GbmaiZfQnqiV3hSB2TzfYXzIwQu/1+BfrHtxUc4P76+BjBH/rdaaw2Q/yD7 Ufy3UP436rZVyP/HCIX8L+T/V6hEZyDl8lYwhtaCwLkOWIRRbMwHHRg2jsHQ gz/YDUAFZ1CS9YPxGPo0mvaIZ9a+Mnr5+kkq6ih2YqBF0lxQFutNYkv0IUjX ZcvR2KqyTJxdzSFg0nmg/RGUUN/tKtDkFIXy8nmAsLRfdkY3skjmV8UvW/4a BP1LdyAUp9QWiioiftuafrWMQjK+mbgyk/ydyaRnkImLSGq970gDScZ/nX38 abD+0+CExDXvF94lmu62CmS7zHadS5dRkznZiKbIKFA1MBfSUQ6hQdIqV7IQ DJf4OGo/iy6t/qx6ocKebLDjww/djApo0MAKs0AnenO4efD2tHt4qOs/25s7 R12uAAm9Tg3DkhiYyiHFcEQEDjNVz1S99qL16k21ZVPtpKn2wk1t3Lmp9m1N tRdtanORejl5vHF9N0RdFikA54Lnn3HaSMZ+SU6S1zxd4CRHp6poIG/mYlCt U1lzJzkG6mVBV9Be7GnVYo7xXnDNQGQNpBToj7x/T11ihyAYgQFKeqZWwDcn ZtcuO3euIEq0FiQEto62ubah4BbBiJhsKWiwyE9w/BXbkSyHI8g4ghol4JDc MiZr6THJDgqMCvyj3cbsqiTFDX7bPNx7t/eGbWz8goK/B2sTFMmsF/iDiAFr 8XyY5CCuUQqLnkr4h9j3Ej0KPQBAbxiuS9h0knReHsdIyGIbsnNiVVNRdgf7 4w8a30wW+9YskqIWyXQ7JD6MC0/UNgzO9mG3e8t8uQ/DfiN4LCD29wH0rLG2 pt5eWqL9/eUKURUuuTKxdm4s57mzUrBMmYTixhIXipUl+WXsfOJTBhdfqyoZ CB6W3kQ+lcoyrDZq9Q6BsO8Aws6CuFfHiZVoZjW4RIPZuf9O7cMEvv7TueOj r/+amKj2f5vF+u8xQ7H+K9Z/Xxxu3/81VC/z9FfpYTzaXJxUQa1gLLeAnVfA 5gUojq9eqsn37KKqkzlihtXSruX5+GHDB4E49cTnBX3a4reNv6moJT7txeSF 0RmpHevVZVIAN0lBkxRsKLbUG0CtSEYOzzYMg7Gp5zCLCJ2nI8CcLKAMuk7/ HOFE3tjD2YhduM64jmRuyQ81/eYJRbKff0404yeaYi/6RJ3fWnJPvCPTbJVm a2nQdLUp4E/HPVSHhzBX4/45Np7v9YoewcUeV+jWUREWe+4gr7Pn0RbfgeYK bubg3uan0CWeTBq1Ir6P3snGhqKsj3hsXJIZRZXPnnXo1+ey+AMrEw6EkvGs nhcQS2ncwM/b4BdHAsY2f0cWs+9YDMpBX0ZxELqLdV/pkrrulr7jjb6l90Sv LNJ/ImtpifdJ5ePlyYbX0eNsirsQcZeyk3kvy77mrVSHGcmSiVokqFfb6lfH ArwgkRpMJd8lfh3xYjNuM0ikBTvYUKh7Jx0Zb6t4W8UTL9Enwu4zAWJZdjDP ZmvZbJnNVtnsjtb92TsUGlVyPpUgeMERERheJBjKBDligNOTDVEcp7ZAgMfZ FMcH4a0Tdf89dUZ4irO0vERNFK26qFQxwhYRUKpSMcccOh7kP0hEBxZnWvev 8+4vyYHLuc6SStXvW1h59EH/89WoPgM5P1MzS/40uJ2xlIWyZlGzpFnw8z0E wC0LhuT46CH1P67/dz+BdjN4iLsfGG7X/+Grdv+zbeH9Tws+Cv3/MUKh/xf6 /xeHO5+v8M07Grdo2guhmz2f74NGfLvP+eSNnRGeC01GLkh4eTIUARnw7ZsK 6cAI5sy7gqGZAu8MQaBIFZcPiTM6C2DszsdScyY1yAFtuD8NIyg4upE7jgjK RT7Ih3eCu60odCluUGHOZDLyAHoccDXqHLevhgEoGdeoY0VuHK0jDPhCeAPK 6/QDCgCxcTBQHimM+e41mwS4AIEqovNw6l9iAtIdj3Zwt9KcALHRnjh0AHuQ pKMAandYDwiIK/SYr+f0LyEHXzrEgGlNoQaTYB3B4Q/swtGIXbkhDA2oiEIo J62EXgpdQAdzuJ+w70TTMth5NbdWTSBdn3uAzbUH0OVMQ7Hv611MqGYAgSoU YZWAy9A7m8IoiMyiZtUSmDjUEL0l1Hmydto3d0YhkP8NZAqvcpCA3shggEeX fnANuYh+aced0yt2bAT8ZMw5V2SSFXaYRhCxO+EUcQTfmLWOiI64ysPXbzgU 3gAU2JrKZUOuPiAck9ruXhNZUUZiXsgraDWHVQWjAU8GwgndcXCFZZJOMJqF BTODy9GB3wqRKrt03QnRjF4BgHP6cYJlA7B0BgOZLwsHa9CHi0o1sW2AnyyW M9mgGP/u4mkIhw31X4CQxuO+MR8LDq4F4KYTyCb0pRmYAIuBieBHQ1A9B7zz BGaiG4x+Ek3G/tJHXZvTCNCBwYW5RS0Igd2rDlW8rKaV5Cweh4+PI1AXrc8G kmZwFTOC+BBEEyBQxbJqpCdB5PFFioXVAINQkFLpvvuMsvR57T7ORfZJfHLo HxUNnEDlAw9XT3K2qKmLCBq0QukIUmapcdbNYcKX2/sBS+44MILjANRbrJOU /vqLVfj34oXIcejiCPl9d529CmFUt2rU0H+4YS+IAL33wF22gvF46tfYZm2r tovcJhjVmLVWZRMgh3ZrpdVuV0FHaDdEtav8JJ9rt3TBDrr344k4h3f5Z9+t lKW9gXv9cXfzn6d7+6+7Ryd8g+gSNH46MquCfPo06ciMWBw++gilyi6q0GVT P67ieMCfKoO/QI1VNvZwrbLY5pBANLsttHe0tQ7IwkCDbvE7Ta/f3TCglVMZ McQ1vPhALOkn4ii+9OkLrm3LuIQvE674BbClD/xD3wBn/OBY0yXM1WVT8lCl PJ1tQO9RxfCtrg7ldoLgEqcrlROSg84yvRwJQ6J+OtZ2fSSVVblMkWeiUGUq I+ht+iJ+ifPZihx7wGDJA8SsDizr2UvEtYKr2yWB/BPAudJhfJEPK70J5EX6 oPU79ZBoFa5ct+h3qkZct2ItF5ARJuYzqupCr4qDeSmoAGq7SPajvOHSE9Ud HycnH7Hyi5OTZMuBSosdEBgpqOWiQ9s7KNjkxMeemARAOEgcGuPji2q+FhY7 i8d41+b63CU9XAwMDIuC5EmORCKWWNw5Ldbz+l7vCHkknW6v3HkDkgTkJ9iQ hHIwa4f3Aw2P71LLsZmAEaaIFTzvDIz35I4LDSt2xd4sagA1AvRPEKUR17lQ tdK7JLlH8ErqSmzrpj8C/sN5cjDhrA00xZgrDmNYGfAjem+MfJJ0VWCWyDAV 97xd0pJo5pwJkH/y5AnW8Qt2RojKK/sfjBK0e8nbKTrsGfKgDrEjLGDxrysr FQQ0J7uWm/pA3kd4B0uOEDWZM5ePoJwCEU4B/kXEoHCRkfRdjCe0zMwhz+VB DxH6C3UJn5x2hddDvInPrpI2TQUZyDnJCUpOEKiJTxFPTBHk1s+eEagTberm IEDSMoNB36VpC8U1LGga2wKV/p1R6RuoqHWORgFSk45Qg+IUIjULjmFDYChi Ce4IoQKIDkOSD2jLE/kpLChAwMMQ3CqEBbmR5iR7Q2qTSLvueBLfVMWqRNxJ wXnfc2FdNYRpOliXTFitm3AFC+QOUwCXxL7QI1xkH5zLDwRLgWnjxkod0moQ S1+AGITYEd6QeTHVOyK4QJZIzJibQeeNgcGBSACpAjJRYHPljKau0P/4LOew Tyd4rYRqcM9A3QB0l4S6g99J+Z4EACocR4PKDORibE2f32Qc0G6AdmcnStYV pMrV2LshFjrn6zY/kCiLJRPebpQgogSIRzJRB8F7AMZ2BJwLOQUxrhnAHAGo SsNJHUgbIT2n5+EeBAI/w9WEG/5AHSo7Ts9j3kUym8VHc4QAEFboTkZOn/e0 0Jyxz+ieDrLCay/im/Gzh7mmk+9QHUiIDkGIoPOAJohVjPEQg+jgdXfneFNR ATTz7xGuKTmhBLiqwB0lcRKgNY76b5hGieu7gBjMxRs28i5xF4CuD7ohv11F skywCNQSSI5JIZppX6LNyzWPEIh8g90d/WLcOknkuxG9ARk7shJ+gCP0I1HV uoQqTyqsjvdyw4ChTiwS0HgQI1mJOJn4rI4aZYe43tk559O4S6X2ttRmA1W8 ugzCCbUSPsd4oaWfRe0fRbQfnFTQ6mU4Cpy4gvdo9LJZZj/iLFbP9B/9xyLV PVP1WaK+UkkeOSxYfIkO0FjSU4ARA2iVfHCci+eBA621QtqN6nK9NZ9zOhIH blY3Uv9ki/gBaIUlGY+rAzzGoEXF0tOfGgP2U7tmD9kfDL//wZ5WmSq4YlXn 9smKdVJRuWagh3kq4uDQGNL5xZRQTVAeEI5J53e09vDVEI3CZyHG9hW3wQ2F 6P7bCSUuc5tC5mpHRnzqoprykq/zCGGxQoOIqrEIFHcgBXoftG0J2pkBqSzm VM4OhbEnQYw8JuWAC1AhzbWNCaEntCpq5Qw1pMBwFWe0skIcZXNAegaW5Qm4 plAaASKk70ZpmnSiwYIKL9VXfaHFtX6lZkqFP0oWRPAL+O3I1dQnXA5w/SkS +pM4S0etHDWcbnJFlvRwEnZc2aGaoNo7rKYXupT3fd3Ke7zAz/+0i8QPcAg4 1/6raSfnf3YD7b/X1gr/L48SivO/4vzvi8Odz/9oE9WwXpBX86QRw73trWaZ WyU2Tdp3W3qawft/zPerbDwW5k9QFf8GmjS/GF4Vq1GhapeFyVYMCOHZm7vx fME741rDs3u9wqphoC82E8MF0T/alTt1wuNviETtWt14jILYfwZi+K54wawV 96vwvIbMrPFUiRtZJxiTnrSEp04bzDBhZ0vjsbKmTpmu803BWcbrOdbqMHcT 0ybRZYbFikpNmernWqrINRYwpemEgcrj0G61Zo0DTRarLdm95lKt3uGat1Jr IUbLrN0k2wCCMqzgMRhuZyhCu12HJS7SJXiuRIXyTj5eaH6DtBxShU4nKj80 BnppG/252EGBHORKmTIQp9GrvNJluN+5rTDGaq19ZkF74U/i2mhWVsxF+fOy lrROyI+53VqGpWZKX02ViDRiYy5LkLy3abMTF0jIYdgCfQ60D6PI97xTSQYZ eJmWak3yaXdWnOLQgh/oYyr2SYxdBVlCKM1Qe5XVkSVWbmUd6R4BgNAfxmRh 0oGCuQmi+klgplLp6B2aGKPE5pMTVzZWrZYCy7miRFyx6Q3NHCjpGCRGleWl bu6T7l0dkJ0FZAgBupxbWVIiYJUsh5YViEoGgaTsLxsGqOy8UDtO1OeSPQEp 8c1J7TCG327FkTSbcoAF3wYgAYOzG37Tk99er6I8RfmpMMpKUgNZ01K3upRv RpW0G02osPEaZQnSLJlmX0Bo3EhNXbjMkpgUDKZ8EqwFewPFStaViZYdpMK7 vV83d969PoW/tOS7p6FcWkb+ycy9MkGs/6T94YNcAZ3r/6uV2H+1Gxau/xpW s1j/PUYo1n/F+u+Lwwz7L1rkSc6iVnjcf5Jcdy22MFFAKowz20VyftRcRJG3 CdNLV8Xw9WmmddSlkKw0Eu4rnla9quHECg+Rh6NpdC7zCodYGQBZN2Ucjrao u2Of5OqfJcOLaOKNgOQwd+QlhXDixKqj9sAz9eoVdg8P9w+VAf/IHbOBq60W uS2O9FB5r7bYZlsSP2nc2IIakdMGUrhmeyKQmsQDYP0n3WLm8v+NG2/Gwfj4 ZuI+/v6v1W7XNfm/Rvu/rXYh/x8jFPK/kP9fHObt/x64oRcMvL7SCjR+A+N6 DrSw7MBvpRVs4Cm1545c+NIp63J6r7vTTaQzirgnMHj98WQJAVRZl98RIa+S 8iQX4OA+yjM6jOyFrnOpzPwW2drQkEWt4qdIeq/iNSL4hOuT22+sEOTtCp6H SjEgsi2wzE32P8lOVd79WWd//BT9IetNrXU1HG9b665YXClJVqWc/+t2nY/u /8NqtFoJ/28S/2+3WwX/f4xQ8P+C/39xuGX9Z1iMk2EbOeXoIatfXZaXZJaW nJ97FfbHH8DOnZ9/ftKr8HtXC2jhRgWp0zQB3dnY6H23Cvg3Dpz/q+Ozb7H/ t9awNP2/Rfc/GvW1gv8/Rij4f8H/vzjcwv+TY/vU7Y0JXbiuSs9OyxPxFJDw 3zSZSOcxc7zhGu8GoBMnvzqu+qgiR1Vxb8NHMKA1w99TX3yO8TbHKX3HQzD+ bbHrG5mbCMZWHjXLcInEDxN5QnKeWOI4JvH0syNcCC1pHZB+GoBDSLv+qeBN df1+Rf7tiEUvR1y73PNWOO3HwhWJRcbTLHbHkyB0wht27o4mMFyYvTf1Rsoe eIhOOZkY0JoqzQ0boJEIh7ZZsSTOSe75E2LdwRlal/4mLGTFTX1M5hYD4uQb L/1TrQFZ1BPQdWWEpp5VsAwXQJRr0W6af4lE8zbFvJdi9Fieuyl2oZKlxx75 7JRVqdHFCEpGXzcrVi50TjM6dF8RDqXhTYA63e4fZxPIcMus05d1joX/JSN1 LFN9kfr5fvjw1uRihEmC1OgsW/dQFbrRdKRtyUovAnTLGymETB4xzaoJWnVG EdpGIzGQIzO6dSE9nHGbG0Fbx2iyNyTiIgOWxHZEOFSj4kvEMyps4E7QawNa DvujG3GLXJIeZa0qGxbsMopSXB5d/QqKT/s2kzAijqXMSVLC63sTMqjyfGXz IbjXhrFzIvsVS0L6R59fnBcc7va8Y54Xp4uA/VKU0+xTJIvcEFk6wmrFiB13 pFWlYa+ZLjzOLeyLwppHEOAT5LXM/TShud6wWc8jX3/uGYh8PCE4B3F8w54g 3REsncCwcd7JM1X9M429876cVaaj0ujvy5f281TUH0sK7MuXVrOSTtYlCVB3 ypRHMnQ1qDqXR79h/DqJzwFQl0pZkPVoJnzVmnfq7svMpCcrWVPni7hb4i8O fcuxS5XhMuWwbGMWH+Q+vaAHfXnZheZbz42v0SbRo/l0IS1zxiacCwnnMoEz zoVzQXAuucWX9If2y0YdTbfH4tN/sjFWfsw04QhKaOTccLEkOYCQY4b/BFQo cWJf1ASvU9wHbTzIGQkwphmsLzH5Sji0YN9cCRgnDtD09LFM929xkHY3VQem HR54IWroZ1spAVSneViXHIwpj/ByeBa9ZmNW/Cc9W/szBL7+p4tir+k+1QNs A8w9/2tp/t/W6P3ftWbx/t+jhGL9X6z/vzjc9v5vmrXkvv+rGWfk2WZIe4YF BEemutQafea7v1TwjfZkCRQyLVDlLyywzt/S4J5SxZWkGe5jlY1r01tnP7U8 /B9P8YRjaVD7+J1cfofIrPJeDc4TmDOvoWr8/8E2gOfd/6zbGv9vI/+3W4X/ /8cJBf8v+P8Xh7n8fybn1+3vJCdemOvdjcFLzsqY4tbZV9zJD5hk2Lb3tHrB lywpZwSzeH26tFeRK+mcqmgVbGTXl3D6i+VykW7ggM/O3ktY3ComioXUf1fQ 5L80FPrqKsA8+/+m1Zbyv1Gv13H9BwpAIf8fIxTyv5D/Xxzmyn9lg5i70Mva HuaeFVvV3Oic4gRa2DUmppNkxyhdAljqW055Ydl/LrDWlRTcVuUXP8W5gSv8 BUwokn+cev6VcBQrT3UXFc6qp1JqDcFNnwSLfpGHx9nXYBQ2dy6pWQyjR0/T 3pSfaYkpJnupKn1p8wMPzx8GmIVQoJmnkFmX7lm5EzPmvdxI4aN7ZZWlhE07 qD3i5NxSWpyFpxAbmm8GppQ5HCX+tInwQfXx4mSFDkSpQ1W9/GgS/+LBJV0V Rv+sSe345URpbfjrI8fjZAM/OjI2ybwhB78kzvqVk6PZVVsPXTVoeDwGXUPh GCNrc42BXOcOx5ig06XhpPr07f7u6U8R/CM7Kok55PedsSsibBXBL+8hDBHB xuhyC1iug16g4atlM7z2HTF8g0hWJOfD25+80588tNuK6N+ZfCqY2x/b8xEQ DttYVU3hzoxaUqBz8gmLMzyMSk0AdXVB9Fi6JMehwV68eG41Grb9mt2nG+/b izm1L9jinxoe/MPideH4rsQs9qsNuhkV5pdrDFt+fI2J38JIz+zMhM7UZtVr zaH+l/20YkclVi+l/sPKkdZlvxHkj3xCeDirYebOyWDNy2BnM+DhsJZFLAhX l7WxN1qs8yFvNh/yBB+ScsRkCFgpsQS8EZQ/QDhGqZ5RXGAmXztZIJ8l80nU eJOhxS6+HZFDa7ul7t5rwzJTJf0IQaSQdWRiHJn+TQ/CLk2EdQMXePxyV5KA SBtyjWe4u3QtVr88aOs/vEG94/pn8flXdgI3d/+33tLWfy06/2sX+7+PEor1 X7H+++Iwd/2nsRZaiAD/Pj8dUUSyuCLGje6tnTuslw42j9+e7nT33hy/Pcqs mcT26JI6vSPo9EiG5qFr1/mU4CfdYQmHTNyIPsH25yXr5UuvIq3opdcmfnUo qSJx5pPs8FafKm9AIpJjY6VicTNYbB3z2M9lpu31Vu622Wv2TyH1imAGTf4f 8idIv/4x8Pzz37VE/lto/9Os1wv7/0cJhfwv5P8Xh7nyX2ctX/0Y2AB+19Pg Jb86rqyTiGdcxEeVW894E9Vg5pPMad1A1fWTV/3Jq6zj3Z8LTRnIqEizz4GV RsDuffxrdtedLgsV4S8XuPw/xGd9OGE8/v1ffudXyn+8Cwbr/0bh/+FRQiH/ C/n/xeE2+Z9mLbn3f+WvkMsmfjaMqMKQ5Ph4p4B5cO7GKB6pBBS/Uj/Suwtc ilaXd08npz5+RPzDgw+e97TnxVVZ66m3mAaSad6iCkiZENlQju2eSeyXZQ5C cUP2CWZRzdXyeLPydMqqKRvim+lKKdFqqgTm2YboI8KMYiP6VHDEYYbQW8j5 UtJfaCzFe7eiNJvEqzXl5v38i7Hhovs9LpV4jtSWTAIkGagN6+XLUTqZD+cf G0k2MzFKJyaH3H7s+VP1vh3M73XDeXEpaafubruUGX5B0IqSJRFrlCrI7ZlV 6eTCryeaIffkvZjJVJYSF9HtpPx/INFPYZ78t5q6/Of+/xrF/v+jhEL+F/L/ i8MM+U/f8H2yW4W/7t+Dm79qLj2ERHLRNd5CS9xbRHHiD1d3hitX81SF5L2l mWJbybIliaxulM0dC882ytawm2WWTSX5ZR0hBaCCI6Cx2I1okyDSZAwiIHcF NBf6C9WYrnBBITNHvgioArfv3//9f3vg8v/IjfcCIq5vYP9lNfDNNyn/m3j/ 226uFf7/HyUU8r+Q/18cblv/J6yFX0GeaKfpEcr+BaSOBiMl001wKL4LgXOn wPm/GLMHqmMO/2ct5PlrLXut3Wq21tqQarWswv7nUUIyb2HOeYHptxuiRl7P jBuji2ic3UlUrbZKvqv5XzM7pO0C613d8cZeHGULonOqaPUdvu2bKcjTjm6i 2B3PSJTGILPK6gYji+SJZmRKmNoPwKw8EA+7m/883dt/3T1iQN71epoBdj/x 1n7r8Z0XtPnPUX6AOm6f/5ZVb7b4/G+011oNsv9vWsX7v48S5IujQnWJxKt6 +Jg6LOMWkOC4+lxdZgN35KLLOnwJFNMjYfujMtZ0uLmeofXcp9xNtBHVk3Xx NWbE4nDqrg4dWPXWpL0J30gYqtakKxXPG+JXfH/8RLwR6/JPesdd5BabF9rq PtmPSHJ5tzxxdEdQ1NfJyp2/ihCCGuRJ9Dzao0iVyOwOLHC+gpftdRjpCwgs c0mzY/TirOcbklwLbv2IAfX8yTReZ47094hO0tiueOR0w2LecMg9p6G+DNq/ sUwg92nSo5vwI8n9w9XYPpa69iINWr3GS0PHikpFp+RXjvQMmuYQIkWXyGsi JhYzEBB1+dLR+U0w5e4sz50r8vYGneL1PFwwkMtKtHT4u/SjystIzDiNpzoY Hb9A9rweFs2yqjxamVuiQYeVf6iWC8DOA5D38jIThpe6fafw5EMGQmKURUH0 l2f0PExWhTO3F5QIiBIKeRi0MIRlHK7QuO89JERYWZErTXLiCcyHD6eVKmzf pTDVnBDKgHe2SAxd3ORyffLWF1y54ciZyGVntnECjSXeOejsUT3jKao0ydAS vm9xZQsLR/nKJ1CWqMJARhKagG/Ph2/fA/49yFg8uE1IKSeoeXSdYUfwL4cL ZbLdxlu+LPP9fFYJjhYHjK9h9cYZmiYyR2dgRC1tv9vpAvwhBzgBds9nTgg5 0fXyWJo+04u46IFyMGRDD1RbPvzvJJu8dvi+gusH07NzNnbHQYi+Kl1fOGgG GJMAanFDQTnSxy4+it1z+pfYAtqoCCIPRSnzYhjsAbJHAUPIWKoLzYpx7yKL BSDJt31IdnE6dJSnaLNBCjVElMOC4k6MziZjheHYGdDjOIhppjECazptlXst PnS/aKWqz0s6grDmvIlnmgzXlaVlMHH9xHkvAhqh8k/JUuHXxgWtNGsa0+AA aSTX2d7+affwEB8CVrLMD3i/VE1hstvdlVnF0PWD6Wgg963ks+i0eWUURAKS JTW0jNLY/7y/J0PClU9HyZiddF9SfgmqVtaFkPd1HLwTjW9BPeSEmfMn2ef9 IAT2MgnoTfbEQXPEW1AmhKmiBEXaizMcSgvJrrmR5p5aPa5K6JSYlObVpAuJ 2iV6hEDG73McCAth8RsLzmBzD+sZIf0i9Je7Rrg9pB0ndGh5wfZpNiAHkymS uGgCIpaGcjCcgIzHZP42Y7LbevR6W7yGLSIkvJosCPJd+N7FYlIrU9WSGTVt A+tTNhLFRW+up6fA0AujWJsDWmY7kzly+4E/SOemkZW6h2RMptNzXTyjg3OO jByBGDe5Uf3tx7BoGt0Qa5TQFDmLQlql9j0qte9TqS0qlc/Kq3xGjaYOhbqC 1JdknabeZABFMl1P+aIwoPNssvVIy+vMQ/ft3tATlJSU1IVBqt+Q9u9UEhoP RY81iiTNCkkVmjKNpMCLpj2UW8nwmmoYwMgh7AOuZ0L3i8qIgokX2uRtGjk7 HmY8YeXUencrGANcdxumWSRWap6Pkw6WBdovW/wCscUTs/OeZ2Aih10VUz2G rggdv+/KuY4PpEvsPRgZOQVln9FYUSXUCgImOgGbUkZ+CkssByacgs1c3Cio sfrLDRX3coNG62gKBbXqEIbsdpiQskFYKTIPILixE9fEWNHclz1uIBinRlIe ACUgidxpwzNTKWh+INUxXYrWOi1hQdoPndgZcZkPQFELcH3ysAJfB7X06GmP zTO1ucCfva+mJUE0tnLi8ji45O6RVU2+21WmSSBmaLTf/SZmEe4dtP3foxj4 XP8BtoDnnf9Y+OYX7v+2GvVWC89/GvVGcf7/KIFvf4Ka3gdJAdLpozrXOOmU ieerpwy0JCMXghg7n4RbkySlqiLJF0rBRb7HIOx/1QO9D1HHnPnfbLTR/rdl teqtplWn9/9sq/D//yiBH//I8WcxPT0sdzZQz+HnQfLUE59fZpZll9WDPa4o gzuo/KXiiBeJSJxwvVG+yozFTz42TtgG+0/56dun1flLyi8NT9+6T6vlpzse VPb0lfuFVT59hQCebtHfPfq7T3+3eQzVtedgXbtnX1rX5gjhHBHmBwT/iNdO 8Zsh1vUPHkM1HvXx7zHl/5XHh4SJj3+3se1PtwL8u0d5tqb4938p9Q1BeEN5 NiOCRt9fEYR/UF2HPYqnmH8R/P+l73sUv0uQjwmHQ4J8eE6YDwjmGdVI399R jUf8L5U9prreEcx/Uh9uEQ6vCKsd+vt2SDnp+2+U85BK7VPOd4TJQUx1Ue1v qcZj6qsDquUV70nCc5NyHqLfrKfbVPaQIG9y/Kmu17y9BOcVteUt1bULZWdu jDzdIqwOeM9Qew/GBIf+dgm3NxR/TPBf3xBkwqpLpY4p578odWd6W13HvIcJ 8w+cAicUQ7VsEpwt+vvqkr5Tu7rUim2K3yVM9qj2nfDp546c2WLKRvyFI1zh OH13IvkDC8IzvLqXrJqyfOJ0/x/Mqks+gFrC/qXOCCCdGAEsN9aqrF1lz6vs RRW0QfgfIiyIacD3ViNBymA31653dp7iNmJVKqr5jXJwrkMsB0RLrV5vv2g+ OONpQj32Gg7dWu1F0+L1vajVLdt6vnjlVr323ILClg0l8RPBrrWxk2ovXmAz rOfw5Tl8ses1q/2iDTXadu3F8xdtyGU3a6BR36WtNmD73GohvOe1+vMWFG7U ay/aIKLhG2Cxhp+tWrNl4/C8gLY9hyrhS/3F8wbU1KzX6m1oYbNZe9FqwXA2 27Xn8LtVx27A4bQAYQDSggwNTGjVnjcBVgsa0mjY9GXtRQMwWGvUWk3Iudaq NQDQ2ota24Yq2nZtDbqiDeVtpJM2FMQPwKUOpZ43aojRc8BxDRF53q6tAdTn kKuO+LywarYN+V5AL4FaAl+gK/ED8lp1IA/s3TomYuOt+lqtaeMnNGMNs8BQ 4HBaMBTPMQEGoE0jZNXaLSxrQ4X4sYb4IIgG1PiiWrYaBBTHrNGuNRr4+VxW A41oIu0/x25q0zdoB6Y8X6vZhNOLeg3rA7xt+mzB+OAn1LO21kICqNewgXYd Rh3HAophCbsOtTyvE4lgslWHPwCibGNv2oBn3aYBxxG1sWtt7DB7raG+IXQY +LIFo2th5fD5oo4UBnpcDXvTopbCIDew7TjGaxQBGNAnoGC3sMlrdo3mN45+ nSIgC5ZZw/Gnrm43aoithSMMtF62ke4azxEx6Mp6gxoJXVe36UsbcuNI2E1s RBOxbrblnxaWamFqi34+xz8vqGHsc7Ekkfr/2Ll0Udl7mDrm6P/tuo3+3yE0 2/XGGt3/r1vF+9+PEn74uqEM/9GRID+Wxf39vPv+8vr739//HfKbF/o5UZRK 225vig8M4p1/BPsAiKJJQV7lM8N+6J2B8gMo2QSALA/6uBGPZgXSpAD3tfvB xBOvP/o3GB1WsdVXHj5YzC0TwkkQ0SkwAPIDfwWSh15cY2wTSvBD82nknLnS SACPfa+htG6YkDVIAGDCJkFgKG0UQDdyU9YGxkE+WS3gy7Q/EMoCHDau9kDd rwwTftAtEx6gLvbW5S9u0nWVM9eH5S7oiqiq0uWGiI+bPI3CrTBvRIc+Xx2X 8uvtnc03RwzU0ZXXpwe/7p7C5867vQ//BDQRgcEU387Gg6fpmG0lmfvsxyVe tsJW9htsZdx8vgZ/YdjO/BV143HDVnGjIJjovy+mY/576AcrURzSnZoVftwG kfjcrh95V+5KMIm9sfc7dUBUnoXDWfmH/KTyr+h0kGF0bbX85nTn3Sv+o7YK /3a83moZog43D/8Fke9P+dXqkedPP9Wc8uYhAoGE8Kq8tVXaYGf9/lcfhNrB YXfr3f6HI6jnxyWBTKVc3tzZWS9pESVgWj8ucZQq8I3aUkkysOVa8ADUuuNF NNvxmlXIgt4FPYOMd0bWv3ptWmPWM3dJa4F2aRV+/F/5FSRtezGdM0GEdhGU kvXriBDBb9vCF3VFlnLpN38hQt0gocTMU6SBdndL/lBP1agSuvPiIOvoijJm vBwFzPz1f2XNEDLQj98eZJz3gmuY8sjV+b1pOpGnE0jJiQABD0QG9BVk19gS cmdxNTh6AIZpDGtpvaT/7pdhhmxt4XTYEqzgx6Xd/dcfdrrwlTEzb9noQwCl /54HysxbzlInwMtEzgOaU6CsEzmHKX8uAC3JmpodAMmImAcrlTk1mQCaETEP WipzWc5FgCO+zoOgspW16YtDqH7NHUAtZ2rWAxwjYh6oVOayxjQAVPJrHhw9 Zw6vAVjZt5DngMwpUNZZloR5B2gmnITbSVDJU12LQNNy5zBLCdNw/74IWLNA OYfnStCmZ9lFYKdK5LBugJ31WjcHdE6BchboXeAZoHThAZB0o/o5gIysX5ub 90eu46+XS+EYtDxUWUhtAYUrLkEsfrKVqabZLQMO33pZWoRHCty8cdM/G3m9 107sPMQJ8Pz331vJ+S/uBVl2Yf/7SKHw//BX9/+wLPdZ1tnqNApXJ24wGbmr QO+riEW0+n71CNaZfVeYOh9uHRn8oHrFrJqFXb+2Wm+s2nWYsOvN5nq9zgAG aIkT+vwRK5K7OOtU5Ct7mcBjPrxmj94iYMCm/ZidJoiS4yO6zRS6UaeERjh7 6sIuRMEqil/thPwh7isCuJIwspQFDnk2cZkfMwgjNijjdYgKIRfeZlIGYxIa 3XaustjDG6Wp1/IAR7GmoyLuJ2c8weuZAF+cV0LjgGh6HYJPW3ZYhgMlpB3W eI1WLf1LmE0APnLpSHY4Cpw4WS5GuPEKNUYTNCFF+J9Z0kO5xx2C/8OK7cDx woMgfgAhMI//NyFS8f8Gvf/aqhf3/x4lFPy/4P85/D/ND75vIWBim0gC8Tom fe95fkfjtwP+mwuAZfHMn55hGTLgA5rAQU3wf61TYz7mW+feQyn/f1uA/9fb Kf6PT8IV/P8xQsH/C/6f5f+KH3zXfF9gaWj+VeL08DcM4o6u4OeydgHhr8XS 7xT4mL92Q+/Kib0r9yE8QM3z/91q1NP8v90u7v8/Sij4f8H/s/zf4AffjwwA +tAwW4efxwG91uCM2CBJoK2RKA5CvIuFXgfigKxufTc8u0GDFP/MlXsmZTxj dyeuj7bekJE//kDbL199KFYzEkxrDUkxKaIGrwfHVqeEj4xz3wN664CEkGLJ MJqap1yxyM0mFjtTY3uJ4J1GnfkAV1iEd6T6wYR78tl+tds3d6oQls1xE64O 7o4ct/2GLzZd3k/w7I4uqOU5eMqpM3J93wkHKxcwSaMEvNHcLnk8mQMI8wTj HvIfs3DsIqOaV5zn8vyzSOy0aYP551EohPwP+pcP5v51nvxfq7d0+Y/vf1mN duH/8VFCIf//6vL/PuKf2MHCcv/hl34ZuQkIKlpJHpnCKw2s2x850TkXUFv4 FSS+27/UX3MSrJ6xYxIUnTSp4SEMqQxbICGm455UHfJBjIJ+BkICwoXkGQXP e9lyWPD8ZgAjA+TWE26vyENfPgyaU2kwAIPPNVz+ngl3VQNvEoz4sVYOHD/w s+1QjYDU2xoyOQ96bg4SvDBPnlE0CkZXs/ufkmeU7E0z7dZKvpqGXqLzzRj7 c+8WCJA6Z9wv8oePF9/hagp7T2rK7XgEfuz0404+IJF6OwiujMwaPkjjPvhu AzFxvGyHKhCYyiZBjBfT5/Xr+7wZsciEeJ9HhYsRYZd0/NxaySFbbCBtcoyj U8EzMmXn8Y+j01wOchf+gSBy+mt+d2HBHA5yJ/6BMHI4yJ35B8LJGbvFhg4L 53CQRfgHFs3hIAvwDyyZw0HuwD9o7LMcZFH+QeOe5SB35x+cBjMc5C78A0Hk cJC78A8axCwHuQf/QEg5HGSxCZHDQRYlwlwOMot/yNUiqBuwXgZlA7WNLk8H Fa8/DUPusQ1X02gN5nPPY+ZKtcTCcTQQpQ93j15rZW8vc8prXaCYyerGge/F QciLcr7moSU9813huwxX/1oHfaaL5H+e9WwR7ha40t896jsj95vd/7Ts1Prf bjYK+99HCcX6/6++/r/P/r/OD76f7f+cI+AE0eQU2P331BPCMbkBSpGQ1e97 2tXOT5PqzYRnPQuD6UT8jRAoG4LWhIcE5G+GLn0mV0aXl7Va3lChTA2ScEWJ s3ByimcNn2hXedcBkvtEhw+f+BMExiY7EIsz1qFzxCjaRWfDfMs5af59JbTg /6AWPdwFoLn3f5rNNP+31gr+/yih4P8F/8/h/4offN/cX6Bp3v1P2Cjnorm3 gKDkKbH5JMoZTc6das9FZoq8VQD/a699Ev4vNnC+wf3/lmVp/N/m9/+L879H CQX/L/h/Pv83+MF3LwY0bGdJg4XEAb5V5WvCgd0iHbQ6/7RCgo/5dgDjv+25 o8E32P9pNNvNFP+3W3Zh//sooeD/Bf/P8n+dH3w/rB/IQ0dM/mT0W3fnVqU+ BuaMb97R0zt4sOENmHx3DX0al1n3dGtn8+gtsPZrNLM9dyLMCIQhHJlLG97f 3TCocV9/8G8AxFGmJ77QcxNdkvz31MMX80IYf+O5Knnj1BkhwWPKA9wpVQ9S 7++cdve6h2/edY9wvtiNxB2zaGmpVNej9j/s7O++klq4lrKzv7W5oyaeraXs 7e9piQ0t5e0rfbI2tZSjd6+7W2833+1RSktLOXi7/6qryqzpuL19p0Fr69D2 d37t7h2LlOc61u91DF5oKa8+HL5LGmSZnbB3vLl1XLL09h93j992D0uW3vKd rf3DQwXdMpuup1l62zcPNt9ppfTGvzd62dJb/97oZktv/sHutjY81vNUUlLQ 0jvgzQcYg1LJrqey49CUbL3pr/YP9yCjndWykqmXf/NKfHpxzq0YupXgcMP4 IcIZ0pSl+QrcEvnw/+hbpjq406ETpWHOB0dJ+IgUTO08yN1jYOXRR33SnHQ4 5O2RcybM9fn8jTGr2HFNeuFPq/IZwXje/oHqmPf+T6PRTul/Vqt4/+dxgpIe zqfEnxdrWNLjPn/LfRT4Z/jCIahrqZxoaoGv43oDfMFMPNSGBY88NAG4Ji2L 9byYP+eHEX3HR50HfU7zBwZHeMko5H78Ta4zg9Pwp9M7Rtzybke7tEAEXbxc Nj/w+f8Oj6G+1fy31tbS6z+rVZz/PE4o1n/F+i+7/hP84PtZ+uXs+hGOuNcn CYS29TAXvkG3t7nbBYUOn2LHw3Xodu5HOXMwD4KkbxagN4JzC9CzzcuT9CXJ fZ5RvgOsFTBwZcv4gHh1eRK6VxwztGbz/MvIvHBHmTvlUkk+BSpKvxfXB76q UONj/v5BD4Dm2X/XLV3/w/ffGmtW4f/jUULB/wv+n+X/WX7wXYuCNLrpE6Cc A58Iz3hwYc8vWeE9KtwoxFU/vn6BKR6aldO9K8NAOBIHRPcqfA0lq+w695a6 eM+MwODD9yPxzLfPf7gwbcxLYtiGTuq3bUSIerJRNh1ipfvtr7GlUYQ7BD7n d4Jg8s3sv62WtZaS/1bTLuT/o4RC/v/V5f89xD9nB9+R/XdG4iOGSspDmASR lKYgSXvT4dBFc7PIU9ZTIqN75YxOuatHcS+7Px5oRfvBeAxkEull/CQLLg/V nXIY+Ui62UWE/ozikw+6mJPuN7n/3bCbaf5v22uF/69HCQX//6vz//us/xJ+ 8D2t+/SbDgebh5u7R6zUzsoGibzpBzhKWwRRJCulzobJFbAf8Mzo+Vfw+wgz ZDLDkhLyGmSquZEXksIolnI4rBej2mi3koecYrSg1Msp+6CkXE4xbndtFjtQ tti8aE4xaJ2BY9pFfro2tci8ynTJVhCEA1gTx26kkExdzIwM7KShU5Q0KrUN G3nj7K2AA7EHixd+vLG4dMMGsLbl1DENuW2WeZyIfhKysBJSwXROKuJBgGwB yP8KfaGN8O2+mTjHoety0jqGb8DSzt2x16edAmdyXslsNVOBsWPSyQ7MARgf iE7IGNqLWbltOrUJrbzNJpltEq7ujGaJOHPEhYs9alkpb7gH3rk7CBPTeSjz WkTR2wXyiXishzcrdIcQKeYYg19u6OJpscwpdCo5kf+MetWfJXCef7C7/YAn APP0v3pDt/9okkfwZnH/41FCof/91fW/e6h/eezge94NyOKr7w345JeGS13j YR7xKhD5pokUxSjNw42NTQB6JchUIdWzQno+vAqVsT7HSIZux42sdBgg9uiz 6OWcAaAiQ8fMxqECOrU5JTelBGWbHJaSpxuo0cgoji4yeXl8JjtincmMkTBv IWs6O4Ie4WXNTIWpWIKbisNsAPAUDe2ptlcAnazu8Tyky9uf0xZVJlNE9Vqm SbfUY9r5M64F4yBT3uwDUSob5uOEltaSOXmZmUmRP1X53wjrfplRacYukYak DfyVzpAin+55L52DCNLQ6DAfxspLrdn5U6hc/yVB6H+S6T3IBuAc/a9pWWn9 z27a7UL/e4xQ6H9/df3vPvt/Gj/4fjYAGZCHhhj8euf3R1PycEn3yIceXs84 R9fv7Aw9tTv9vhtFwlbL1O5IlQIqKevDwYdR3uN7f6pqw3FCfdAbModKUlVo gtJzgQRBEXTQ/KTG3mGGnheTH3yqgQ9wTJZnWBWqCW6fHMnLiq690aiMBIlV ALJ7H3Z2YNJIgzNy0+cO6BlI0QxQF2LPd6OHMCfzqFNd9lR/BPSpFp9ZGuiJ 2ZtD6aLZ2KPzIDrPRqtHiAwImmcqE0boXLqHUyDtm5xUUHZy23IcepM8cJkH 8PTEvOXR05xViaQeWozwna+0miU0sj33Ghga0kKKQJM9s8xdHV4SmeHMMl29 mkT/24f5wWvj+nEk7y+Rw6tIA6BGxtwS/C0YOJdscu6tTiKPNiKXV5ksoz04 qpfZpLcf5HOpAlFVSD7ExExMIVpzGposzfSimg8svehe4K8kzXNV41DLD1QH JMA1iGlaEl3NoxmPR8MXx8BDozF9J3tHtBgYwoR2OtFMJllDicdhEY4sxMEl lGk0a3c6ijGBjQPgIHK8zaLpBwtFUYxaSRqcopWcNQAV2t2mm2dxGICiFIMs HQfQTrlwkCNSrBf+pIHL/CMa4uBh1P/59j/1Rlr/t9uF/4dHCYX+X+j/Wf0/ 4Qffj/qffQVEYkn6lTizjijOHZxy651IPP+7p35mcw5D5wzdPai8eoQ4GlWZ SWqrnPJXFigey57SNqHMa8R8ZhL5byw5Bf8/d0ejb2b/2Wjr/L9V2H8+Yij4 /1+d/9+H/XN28D0f+RGKCePHS0dOeIqmm5q1C0ZdOSP5BDz85KwXyxYrFgqS /6sdmAeoY67/55ad4v92o3j/93FCwf//6vz/Xvp/wg++pwVAjhQQeKr7Htyj f+LT2T/DextEQMLinj9CC7HBaEq3Q+3lg3ernyYsuQcpHYWiqJAboH9RcSH5 f3r3/GvWMff9l1Za/wf2bxX8/zFCwf8L/p/H/9P84DsXAya6pv3/nDt78rVE lBJ4TsXPaeGvSjB9EybGETOuhJEM8Z04Wze/qw+E6Dr9c3X9UH9VRtqF6OYg wj1McqMLj5466iYXEA6QN+sH3ogOpcTRTOYQK1+ACf6vn49+dR4zf/8/o/+v tYr7P48SyoUAKARAjgDQGcL3zfsTTLUnwPxYHNonzHfm6TuD3KF6s8twFkOs HB95Nc3+PP7uq8nbCdSVM5rmvpELXJtO+vmqQrsq8I3XFZL/S1u5x/f/b7Vt w/9/m/h/u7D/eZRQsP+C/eex/4QffE/cP8v8JZ7E++VrzssZE2QK5KsrsQru azbJKyy+DvD+ZcSe5L0PDRDP8kCaEPkjjX3XpztqK+L59IHQ++k9gid0AS0C 4urTywT5dSFF5j0N/hriHbKU5c9HUtCsPQjPvMLKxmQgIESpYvyqbE6xTUrw eh4Std5PS9gq1nNgElVSwNIWwRr+0i54/kPXAAjvuJ2S3ExfDNzF229c5GJ/ DnsrZGqT2HxHOT7fAe083/R00VDgIoy+sLteBaEPazQAByOoLsuJR7iXr4LR qTscJn4CgMvAb3GzF1KnY6OPS8tvTiN3ZBbACDYJgBd6v/Pmk8d5L1UydAbe aS/0Z1TGEYU8XsayveSr3c8kzFgPs7Sb/JJP9Ny5pbBwcCeoUBi2ldCOPKfe lH0+2ZVJM/TcRbEcN382vJx2IN9mjgIs8SoJoyv9UVVxATtKtby07Pla3pz1 u95YP/BpCHnmAwkSWihTND+Aei17quQtzQJhEbmzAGzmQ8jBOU0YNGISbY73 LYjqGTWFWiuSwUzrknSJBCM+LoJL8sxbgmVqrhMJoJEpQ4Aq2+aMfKpO8kUx HQXjnnzdweA43T4kZgitlC21xHl9wv5kXgmKHefAIiu9IHZGwHBGwdRAIgdA D2Z2HtWjsvHmFYuC0RVnG0tZJnKD1g8k4Oj5m8OpJ5ovTCmjeDoc6tNsma6L C/rt6BiLEmq20NsaNHt5J3iReG+H/P3jiwCjILjEih3DhJYt7+VUkSYRUQmo WgJ8nohEOHTvW+seiaq8zy7eb55VPI8tCt4KEokrhOQgIxfAeQ86Nz1zOZ0A ryI1ELt9AgokAAvY21crWGDGQGfRSSiFDHnF0J33QALLfUlySyJ0iKSL0QA0 1TGCarD0WeIsxrz1L16izpB1SW9y7sRAsPmtyhbA/Bn0UTGWDeRLhBVtiW5c Po9SbQ3iWW3lgL6kuQQhO30F5PwWZ8tgEV6Cc6aIFiEDDyasKxlSYqm8om06 G9PXueq5o4sUhryYMQ1KBjjT8mEuy5IQc0hx0YkBfZAHI6FmH/lTQtFvsbmT 86CHGk3CCpNGHp0fplUQrsdzY4yfTXOMtE6abSHW5eaMKsXn9VFJa1q2cNIw niiadaQ48xC98sxm8stdZOI57esmzH0hWZMFo6GGiQKzV9MQjT9uw6g3zXgO Ki1SEgVWpmiCRaZ8xr4nFxu+1ZaBSSUTz0HoctnkDZBB6mIp8ydNB0UfBKQb pPXPKPGPNEO2Z9BKWgppZjN3XN93wsHKexCRkTcdp+bx6GLGk1miHKNyt/d8 LgyF0SxAJK/82KHXC7RhkEsPUmIoXTgh4O0TRQw4xy5yOWRXInqFnubCiCcp kDFl1SHCb+KRJmYwOufc+hHXnTDBcSUGsxytIIMkv6c74Cixbh/L5alNHOBw 5JyJGtAiSmt4zuKLWAZ3H6KxwUlOOVORRGaReB2hNs4upGiWG21x7UhfTZAE UN7hPN3ij/yNGYqdITdwryy7EgZV732Orscl5xcqe6mqQOd7n6dXpvbmF9T6 3ueofQrrRfQ+BcDu3B9AdqbdTULmwsjT996mVaXbxlbDMUcx4I28v2agw0z6 7qvAzGpvd+nLPBiztA00ZpSzBLSwBHWcmzS35HQsoT2kiE38r2B5msaUd0Uo kTg52fvuDkLaC3z6tEFq6+6FtH7A9iEDE/vHPdzJMlhHN0bcZ637Yk4kOSyn GzlXcjUfuldeMI1uyT3oztoLO05eLh14QyVhhbTUHjPla3jB12N3DJPXQQ+L YiUvttS0BGx0dBNBjC4RwnE0EHgfwlesKRHsym2jXmJT35AQ3JVvLcj7DzKj uTkz0PZk9M4wdw0gl9y4lfsKuAk6iUWF1+LlSbl/maDoUDb5mKVEYuvgw7E3 lr0ymbIYfsEKB5snu3Sk7YcMvU/I1Xn+HeTv0G0Qia9aakoJz8xlaTa/kKg5 ZYC2l/HFvjhbiKJDR/qXysxYSk+2l/lyVxZRW8um+KQyQ+HESMtN73byGh0/ 2SYo+RQlx9ZwiJVCT9Vx3D2bKqdbnBwoQqN7YeVcWsZ1ZJRVlHW/objSjGix iVvL5FlCOaYcTEnF0XyLKgvq0vIA9B/U2a+kp7CEMhCSliz4wOugT5Iz2YtJ 9tCjaQisVDTqiP+AkYhTS2CRjezNeN7fzj3QY/mG1yDABScWY7B+DLjc+B+t s7F4ZmNPwGSJpzFEYYLkZG5ML0c8UnryxN1zrbxIZXgSQmcixHVxp3rKl8EO kPUwPk/UGY1EeWGtXdLR6ZQfnEQaPqItFJPe+dPyqVYIlwXmbreqEv2tEpBt fALWEZ1JZZQbkQCPREZuFotlH/KdClRMvvo2uAYg/o0q5HBMJjRJOaFRTRzV QxeZSRBKZMUmduj2EwxVHt45K6CWqR10kDievNOgoQblZ25rpzff8figf5rr +E7Uq9+mwxz7XUNAGBvsogF7gXsg+M5E2yzfcqpsr8q2quwtfvRyTwQ+a6dt 3/rmxF8jiPsfY6DsB3sEYO77b2vk/8teazWa9baF97/tVnH/41FCcf+juP+R c//D5Aff0xWQ7B0QHdeFff6XlOP3KM81LPf4TjKH3wcEDZCMSymvfCEUR9Lw Wa7yemItoVJRQbzrOwCRVFDT2yrkfz7jmp/8z/NyJdVE4axeayT9LJkO+LXO wiz0k2cBiav3byF0/2IhR/5HX1sBmP/+dyN1/7PRLPy/P04o5H8h//9U8l8+ ArR3erS7ubNzuru/c1Rq1etzFIOINIPtdztdtjycyA0LcuM5SfZ++GHHKFkB ++qdkl0lyX3KkojYSOXRdh2wbk36+2zsjoPwJtkuWl6WqfrLdZnHVtICOPrK Eljwf9qReCgHQHPtf630/X+r3Sz8/zxKKPh/wf9z+L/gB98P4wfSkEjB121v RJe5zukWFd347TtXXoyH9yNobCSZeR9vDYjhwkGVN2rLTHsPzImioO/R7QJx eXjW22G1r9yqvOUstVI3Z7j62DhJPFqEwixNvFNHAoLvNt9LMAj+r848Ht/+ i9lrdjvh/806t/+tF/z/MULB/wv+n8P/NX7w/cgA4JZS+39/iqr86fG/DrpH pdLzTMLR5u4BqPo8lOrp5Ffdo+PuXvfwzb8w2cqUPt48PD7YP4A0O532Bstt Hr/b3zuC5Eam6NZht7snabbUTKcD1FfvNo9keisPs8NdTC+tpdO29ve29w93 ZeXATXPEhxq5ZCs0it2JsmYmoyr8xddCXoCLmeij3qEn8nFsTCKfeqn0j7ub /9x5t9eFjOlLAn7g4nQCkfXRPhE22MpqrT+Ng+GQDsNpyQX0F6OIVTZp2luU ACh7Mchb+cV7ZlX5py0+G3jsGLF+MPVRhIueVSY6y8sTsd6TDuvVtZ2e07/s 4VyXdyRKsqisUwIg4PIOw9QXE1nerMAXpNBRFL3wldpaJVs4bWv1yg2dM5dJ EznILpd5auQ6pcfbZRXyny7lfKv1X6PVSsl/q90u/H88SijkfyH/c+S/4Aff lezPiDpCMhFzuA2X2VzjCyTKWZxd5QQ+5ng/8Yi69Rus/6y2ZWfWf1bh//VR QsH/C/6f5f86P/iuZUCCKMoBXWvHlH0f2D/etj6NUb0O48SPk4jddR0/E9n1 B1BOVIFxV2i+dvT6VMDIS+KA8lII2g/ekMHy7dfuP9kff7C3B6cf6Mv+0TZ+ HB3Y+AFLqg//VMhooH5HesQG/S98AWCuP/CGKNiS5t9XuAn+H7oPdvqzAP+v 6/t/eP8P/hbn/48SCv5f8P8c/k/84Pvm/ICi0v3FYUhipCmv8sn9pJ43RNPk xKqEIvoO3mHH7Si06khu8aPhmQCG5rGeP+UZheWouDsII9Y/zxhP8GhhEqAj y5bhh99HJKQpF8ZSZE5mDgd32l7t770+6TCFuV4BiADI/QUrG8n/k2cwvyJz EWHu+w/1tRT/t1vN4v23RwkF/y/4fx7/T/jBdy4FJKLJPlBMD0BwR62R97v+ IJDuhFV7YPW/eGOIj/mvTug5vQe4+kthjv/XRqOt7//YyP+bzWL//1FCwf8L /p/l/xo/+H7YP1CHhhf8eq1fzcLz3HN82Fu7u3Uls9PIX5NhdB+QhbEHwhuM YPh7N2XmqIxon+ucueFXHwXt3vKvm4dAzaVSs53E7e7/Ku4yWA39LsPRu91T yG9cYOjudSlOv7WAF6ExTr+tsLX5K8U1slJTdaP+cir0AbdwYnyFIjORtVOk mTdhzoEb9cXSR+XDOJCpyiqeJDDkRX88usm9UYgSjeyIbAcd+eRl197D4EE9 7Ed9Skb3wTBOBpTfCgh1VxVs/8oNR2RCjfkxcP8OVCDgiYY3A+UETN5k2KKL DIaTRAbrlq3kfgN5P1kRLnMQA6fP3Zhkg1kB93fn9YElxq5H7h8lliIKUJOZ Ev8Js0M+ePRQJrzkS/Ai6h7gRRCVvAuApx/B4tY/i887NJCyCkpCMqQ0trSL rjgqt0CfUUX3U380RU5zSM4E+PUK1pIVuDJZOhuYX4H0qLALAPBV3oQ2jvD2 DLCM+Np1dafRLHLGE7qFOgfkIT3QcgTMXaz9k0i5TRG56uLKXERFFxxr/lYS sJs9etbMTXtdCac+W/qHO7ry/Jzuzgf7BjuPaERC/xV4JwgS9puDLpyFdw3a wBCe6uYh/evgt13nk9yHsepM+ulJDCeot9EXwXQWPLN392h8JEgMxyBwQ5q6 Q/Ke8AnFdjztcRzVqC1t+mdRXKtoIAWW3FOdnOMKrja7uZvBaUge7ZaO/h0y AhYG7jjdvRpcpfkIqlX92lW+nAW9Qkco25GlWaA1yNxrLsdbw3jX80EYJpgq noZt4D5ucsBrcDd93yWXWKozcNhgnnX52Pen8YrqE5l3oWFTkNXwCZz3TFJY RUKmr4wUuBl16DhzXg7NPIRZqs+Nfp/HQy/HKNb6/VU/dC9ub4UGeRvm/cic dwI0paTnXAJqBi0bpHHlbnJH58noOZ9o9MgHORLH2HR6ftjdAb2hC8wDZOx5 Th8fTZxrX+sE0N0Y/uRuUSCN/KEiKUg2r2JT3aAzSE5T2+ixJYfUIvffU3Lc NMQMEnrsfooVFeYThGhtBq7ohLvD1chB+a/XydgktjPkZtxzPBLam01dj5xN xm+SYrk0zEdr4vUvk3e/+mEQRSuoZ8wkhy0nDG9QTdkWegNH2YaZd+CG6EwH 7xByl0jKdz4ZPYxc51LVBK3o3/TTVPH60NV8tCSdceiu6L5bhiHv8hus5+g1 uVHjPUPCZMViroCs1IvfgLhCQyQjWMXSrjEZlz49F7h1STZatl8AOYIFmwtL uv6lb75IIOUx965MTvBkLv6ognA7BDpXDCrwzAq4p8otQfPZCrZNz1TkBEo5 t5wZsq4P9w4Au05JKbIAvlHXiQO68gJtDxP6mEUaec3YeZ/1hKg3w/T5mXJ/ rIdZ/bT9ascZ9wZOnie6pJ9eeU7ERpSRugrPvafK5gYH2xFvR8xqCLCOI811 EwbFTbi7psW0uDzQzqejjCsmrQmKu9y3HqnicR9FRy56suKzStWDD9rJRwIj lYELGxJVujerOe15HcRi8ZJagYj2iEQ+ECNvjIoqgie/YQHfEpHT5Lb2HMOc ckdGS4x63vWDyMELzAgzyctXPool6r7A8tvTzTrKMupJ1AwpnxMHqrBu9M78 sZtdneRMeVR14tfulZdqk6hHpVAngfIYiQatyOcVdRVydr8dTXu7W6jR5NAb XmyDdPRgByJ4d0ssLFbJCxzXcGaEPHrrz6xHpHP/XHepJqceLJNXjeo35cBO XyNBpSAiZs8jTRgdY18rCbYZowoVi5c4McgYdC479LgT1CDEVUlCcNnFk1YB YogMJtMI2VPemTdgvWBww8bIdLxILBOoRUPpAW6gNdSs4M3mbl/osJkuarbk 8hFmwi6qKyjZR4GuW6iFpcH6NV3g7cHtfP5IcHao4jzrTf2WMSgb2qfUNKKP yW7USYdbvl8hk/6Eqldf+WJNFsxXLm0cmuVUQUyDrv3EFQXaQeKF9siuRCuE Z917m7vdk07yxA85WlUbUEb5127UTxf/bf/w9ezipc/JBuJ/85mPHsT5jwfc zRntOuhb8avfBJtn/2me/9vk/8Uu7P8fJRTnP8X5T875T4offF+HQAZy62Wm omBwKY4/16ZZ7UOWg1932Tq5dMVvDu4LkuoPA9Kw2XkQxdwhqzMaBddAFyvc ISv6ZPWZN4YuG0utQ1IpLm4gGrL2PHwNoR8MhCCOvUHEX8kjIKiJx1+5I9AY 1hviqc4pNOgUfvj90XTgsqe12uoBkNNo5I5WJ1fjRu38aXKmBELy7f7R8REr NWx5kTh7PmT0MB4SGVUJ09G+E/fPT9FG1Hwe18fOTEXhDpUZdR2Elx8lNidm GnafniavxkFrELbnDwO2jN8mnZJRcHwDRU1YYwd0tpCik3vTZgP/m3WBTSD3 /ir9NZ0+9b9eHXPk/1qzsab8v7Zxk9JqWmuF/c+jhEL+/9fIf0M6P4BjNiWA iJuQ1EEGnOUtbMl0acaWo5uoStx65Dux/NYbVMr/KZslybCzQ3C93k3sRht1 yF69YCIOpXefbbB6R4nH16/elIeTEJKHbCmKYckbV59CJPvlF8TNgL9U+T// aUVJirKHRW6ilV/QuJS9NJzOVfiNQ1mjyoaXC8vuKHLNdL0oZaEC6DtOoIt1 LfGfS+lGV+iNCICzxOFVAS3vd/c0ruBHMEwXqFQqGxt7H3Z2OJLdMDwOcRdu 6Wm6xU+r+x+O97d3u7vQ7FLogr7ksyU97nOZ9zR7tiGrX55VK2Qv457JkrdR 7zDvJc8O3549E90lGvnRO8lpZ4WNRTNva57eOISZal6K2MwG5rYQmqg0BtXU 3Jo7ZaMJK78It8DYFn45JtsEsS0h+wyzPUQLStoo5VaZh7s3+FNiLrwmJ7hr E0QiT5xEA2NOiofAHWvUEV/mmNN8uID5cPESc3QuaC4QAnmN+nhxctuYtGaO BkG8U5vyG4WtmtOyynIrOyjCJzUiT1e58oakN5BDAlkefjx6g2W9ulyUI4Xw fIwfg4YMnBEpYqoJFTEko94APmfQETVKUNGMgbBnjcJXpSGtWVp1y7Zods6c prFQd/Vvn9My22PO6gS1vEGBuT1zUKh1YlAU5pn2NczGPebQiBqXG7ILYIBI Q1FeahF14VyjQmpORql6RUCpTeRMkQ5n+HNf6+ynAahWVV5xJafwLN+5dC4+ CpwB643wBEtBonoqnfupebAmGaCS/8TQ92SHHR9+6ELs5//ifYEiFKEIRShC EYpQhCIUoQhFKEIRilCEIhShCEUowp87/D8iW9o+ADACAA== ---1060964174-1454727082-929096045=:16828-- From chemistry-request@server.ccl.net Fri Jun 11 13:35:56 1999 Received: from nucleus.harvard.edu (nucleus.harvard.edu [140.247.90.196]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id NAA20307 for ; Fri, 11 Jun 1999 13:35:56 -0400 Received: from localhost (daizadeh@localhost) by nucleus.harvard.edu (8.9.3/8.9.3) with ESMTP id NAA00888; Fri, 11 Jun 1999 13:29:33 -0400 (EDT) Date: Fri, 11 Jun 1999 13:29:33 -0400 (EDT) From: Iraj Daizadeh To: chemistry@ccl.net, carl@fallingrain.com, bio-info@net.bio.net, info@ncbi.nlm.nih.gov Subject: HumanDna/Time... Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi, Would someone be so kind as to email me the data for the number of sequenced nts for human as a function of genbank release. This is for a talk. Thanks, Iraj Iraj Daizadeh, Ph.D. Harvard University Department of Cellular and Molecular Biology The Biological Laboratories 16 Divinity Avenue Cambridge, MA 02138 Phone: (617) 495-0783 (617) 495-0560 Fax: (617) 496-4313 Email: daizadeh@fas.harvard.edu WebPage: http://mcb.harvard.edu/gilbert/daizadeh From chemistry-request@server.ccl.net Fri Jun 11 14:45:47 1999 Received: from terra.ifi.unicamp.br (terra.ifi.unicamp.br [143.106.6.20]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id OAA21273 for ; Fri, 11 Jun 1999 14:45:41 -0400 Received: from lua.ifi.unicamp.br (lua [143.106.6.13]) by terra.ifi.unicamp.br (8.9.3/8.9.3) with ESMTP id PAA04294; Fri, 11 Jun 1999 15:24:54 -0300 (BSC) Received: (from daemon@localhost) by lua.ifi.unicamp.br (8.9.3/8.9.3) id PAA11374; Fri, 11 Jun 1999 15:23:32 -0300 (BSC) Received: from gauss(143.106.72.139), claiming to be "ifi.unicamp.br" via SMTP by lua.ifi.unicamp.br, id smtpdaaldDa; Fri Jun 11 15:23:27 1999 Message-ID: <37615503.BB60C8F5@ifi.unicamp.br> Date: Fri, 11 Jun 1999 15:27:16 -0300 From: =?iso-8859-1?Q?M=E1rcio?= Cyrillo Organization: unicamp X-Mailer: Mozilla 4.6 [en] (Win98; I) X-Accept-Language: en MIME-Version: 1.0 To: Computational Chemistry List , " =?iso-8859-1?Q?M=E1rcio?= Cyrillo" , " henrik.konschin" , siviglia0 , burton , "Bendale, Rajiv" , "John.Simmie" , stortz , sgaray , Mairim Sosa , "jerome f. baker" Subject: Chem2Pac available Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Dear users, I am glad to announce the availability of a new free program for Windows environment: Chem2Pac. This special text editor is intended to be a computational chemistry integrator, having MOPAC, RasMol, Babel, etc. inside it. For those who like to use Windows System (sorry, no Linux version available), the program will soon show its usefulness. Please refer to http://www.ifi.unicamp.br/gsonm/chem2pac for more information. Best wishes, Marcio Cyrillo Chem2Pac and HyperSpin developer. From chemistry-request@server.ccl.net Fri Jun 11 18:12:57 1999 Received: from nucleus.harvard.edu (nucleus.harvard.edu [140.247.90.196]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id SAA23212 for ; Fri, 11 Jun 1999 18:12:57 -0400 Received: from localhost (daizadeh@localhost) by nucleus.harvard.edu (8.9.3/8.9.3) with ESMTP id SAA01255; Fri, 11 Jun 1999 18:06:13 -0400 (EDT) Date: Fri, 11 Jun 1999 18:06:12 -0400 (EDT) From: Iraj Daizadeh To: Monica Romiti cc: chemistry@ccl.net, bio-info@net.bio.net, carl@fallingrain.com, info@ncbi.nlm.nih.gov Subject: Re: HumanDna/Time... In-Reply-To: <199906112120.RAA29399@focus.nlm.nih.gov> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi, Monica's solution is perfect... It is quite interesting...It seems as though for the entire genbank release since its conception in 1982 that the database has been growing (number of sequenced bases) by about an order of magnitude per 4-5 years...for human this is roughly of the same order of magnitude ... I have attached the data for human for your convenience...The rest can be obtained from the april 111 release in section 2.2.8. (see below for web address)...Thank Monica...Iraj. Dec release/year bases 93 31625281 94 44354415 95 154817348 96 291085726 97 551650819 98 1084135266 99 1304754476 Iraj... On Fri, 11 Jun 1999, Monica Romiti wrote: > Dear Iraj, > > We do not have compiled statistics for this. You will need to > look at section 2.2.7 for each release and compile your own > statistics. It is true for other releases-- I checked several of > them before I emailed you. > > information for *each* separate release is from our anonymous > ftp site: ftp://ncbi.nlm.nih.gov/genbank/release.notes/ > > and for the current release: ftp://ncbi.nlm.nih.gov/genbank/ ---> gbrel.txt > > > Regards, > > Monica at NCBI User Services > > > > > > On Fri, 11 Jun 1999, Monica Romiti wrote: > > > > > > > Dear Colleague, > > > > > > > > You can find the information for each release from our anonymous > > > > ftp site: ftp://ncbi.nlm.nih.gov/genbank/release.notes/ > > > > > > > > and for the current release: ftp://ncbi.nlm.nih.gov/genbank/ ---> > gbrel.txt > > > > > > > > Regards, > > > > > > > > Monica L. Romiti > > > > GenBank User Services > > > > > > > > ------------- Begin Forwarded Message ------------- > > > > > > > > Date: Fri, 11 Jun 1999 13:29:33 -0400 (EDT) > > > > From: Iraj Daizadeh > > > > To: chemistry@ccl.net, carl@fallingrain.com, bio-info@net.bio.net, > > > > info@ncbi.nlm.nih.gov > > > > Subject: HumanDna/Time... > > > > MIME-Version: 1.0 > > > > > > > > > > > > Hi, > > > > > > > > Would someone be so kind as to email me > > > > the data for the number of sequenced nts for human as > > > > a function of genbank release. This is for a talk. Thanks, Iraj > > > > > > > > Iraj Daizadeh, Ph.D. > > > > Harvard University > > > > Department of Cellular and Molecular Biology > > > > The Biological Laboratories > > > > 16 Divinity Avenue > > > > Cambridge, MA 02138 > > > > Phone: (617) 495-0783 > > > > (617) 495-0560 > > > > Fax: (617) 496-4313 > > > > Email: daizadeh@fas.harvard.edu > > > > WebPage: http://mcb.harvard.edu/gilbert/daizadeh > > > > > > > > > > > > > > > > ------------- End Forwarded Message ------------- > > > > > > > > > > > > > > > > ------------- End Forwarded Message ------------- > > > > > > > > > > > > > > > > > > > > > > ------------- End Forwarded Message ------------- > > > From chemistry-request@server.ccl.net Fri Jun 11 18:18:32 1999 Received: from gateway.bioreason.com (bioreason.com [207.108.245.3]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id SAA23330 for ; Fri, 11 Jun 1999 18:18:31 -0400 Received: from bioreason.com (val [192.168.1.51]) by gateway.bioreason.com (8.9.3/8.9.3) with ESMTP id QAA20153 for ; Fri, 11 Jun 1999 16:03:48 -0600 Sender: dalke@bioreason.com Message-ID: <3761893E.EBD856ED@bioreason.com> Date: Fri, 11 Jun 1999 16:10:06 -0600 From: Andrew Dalke Organization: Bioreason, Inc. X-Mailer: Mozilla 4.05C-SGI [en] (X11; U; IRIX64 6.5 IP30) MIME-Version: 1.0 To: chemistry@ccl.net Subject: Re: CCL:maximum common substructure - the code References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Jan Pedersen : > Please find attached Graph.tar containing the necessary code > for maximum common sub-structure identification using Bron & > Kerbosh. That reminded me, in addition to the graph search MCS info at http://starship.python.net/crew/dalke/MCS/ I just released the core library routines for both Bron-Kerbosch and the DIMACS challenge at http://starship.python.net/crew/dalke/clique/ , licensed under the LGPL. The B-K implementation is based directly on the paper, and both algorithms have been changed to be more "library-like", that is, pure functional and with dynamic memory allocation, error checking and no use of global variables. Included in the code is a simple driver used to test that the system compiles and works, but you'll have to write the adapter code needed to make it useful. I found the dfmax code was faster for what we needed than B-K, partially because it doesn't try to find all of the cliques. The only numbers I can find at the momement have it 3 times faster, but that was for randomly generated graphs, and not ones which derived from chemical structures. I would be interested to know if others (Jan?) find this to hold true. Still, I don't think clique detection is the best way to find the MCS between two compounds. I've sent email to Jan Pedersen to continue this discussion off CCL. Andrew Dalke dalke@bioreason.com