Minutes of the Working Group 2 Meeting Wednesday September 29th 1999, York University Present: Eleanor Dodson (Chairperson) York Jan White Sheffield Alexei Vagin York Misha Isupov Exeter Ian Tickle Birkbeck Tadeusz Skarzynski Glaxo Wellcome Liz Potterton York Julie Wilson York Jonathan Grimes Oxford Sheila Gover Oxford Sue Bailey Daresbury/CCP4 Alun Ashton Daresbury/CCP4 Peter Briggs Daresbury/CCP4 Martyn Winn Daresbury/CCP4 David Brown Daresbury/CCP4 Apologies: Phil Evans, Helen Saibil, Christine Cardin, Lindsey Sawyer, Bart Hazes and Garib Murshudov. 1.Minutes of the meeting of 5th May 99 and matters arising. The minutes were accepted and the following items were discussed a) Seminars for program developers - in reply to a question from Martyn, Eleanor said that Garib had been given the responsibility for arranging the seminars. Liz suggested that time could be made available at the Study weekend for a seminar on programm ing the interface. Eleanor asked Liz to take on the organisation and put the details on e-mail. b) IMAGEcif - in the absence of Phil an update report was not available. c) Harvesting - Martyn reported that as a result of Kim's work Harvesting would be introduced in the next release. 2.Study Weekend 2000 Julie and Jonathan advised that a programme had been devised but it was proving difficult to find a speaker for the introductory talk. Concern was also expressed about the potential number of US based speakers. Julie said a visa would be needed by the Rus sian speaker. Alun said that the Ebor suite at York racecourse was being looked at as a possible venue for the conference dinner. A disco was also being considered but was only agreed to if a 'quiet' area could be provided. Commercial sponsorship was not considered feasible at this stage. A sentence would be included in acknowledgement letters to advise bursary recipients that they would be responsible for the unrecoverable costs if they failed to attend. The scientific organisers would be responsible for bringing together the speakers papers at and after the meeting. DL would provide an admin service for the circulation of manuscripts to referees etc. Letters to referees must specify that the papers are part of the conference proceedings. 3. Progress with NT Port Alun reported that good progress has been made. The low level CCP4 libraries needed some alterations and additions to work under NT which are mostly done now, and problems with setting up environment variables have been solved. libccp4 needed to be static not dynamic. A large number of programs have been compiled and appear to be running ok, although there are still problems with s/r CCPALC1 (dynamic memory allocation), and binsortint.c (which uses unix commands eg fork, wait etc which are not supported under NT). Harvesting and ccif are not yet supported in the NT version. Liz and Alun have also made a lot of progress with the GUI under NT, and the NT version of ccp4i was demoed at the IUCr congress in August. There were suggestions that someone could look into alternatives for sorting and memory allocation. The discussion went into a minor sidetrack concerning CCP4's ability to advise on PC specifications. It was pointed out that rapid changes in PCs meant this was not viable, and that the bulletin board was the best place for such enquiries (where there is also access to a large and knowledgeable user community). 4. GUI Update Liz reported that the GUI had been downloaded by 220 sites from York, and that it had been demoed at various workshops and venues since the last WG2 meeting. New features in the next release will include: interface being released as part of the main suite, new task interfaces (including MOLREP and NCS for REFMAC) and changes to loggraph. Also some non-CCP4 programs (SHELX and Modeller) had been interfaced. CCP4i seems to be working on NT. Necessary changes included: removing unix specific code; interprocess communications now using sockets rather than the "send" command; and using directory aliases throughout (rather than full pathnames). Some issues remain under NT, e.g. problems with installation, environment variables, and definition of directory aliases. Liz also gave details of a 2D Molecular Sketcher, which will act as an interface to Alexei Vagin's new dictionary program and provide a graphical way to access and alter dictionary entries. 5. Changes to loggraph and prototype Tcl/Tk map viewer Peter Briggs presented a summary of changes to the loggraph program. New functions allow adding annotation to graphs, and displaying the data in tabular format. Suggestions for additional features included spawning new viewers for new graphs and overlaying different graphs. It was mentioned that there is a problem with BLT/loggraph on DEC Alphas. Details were also presented of a proposed viewer for map sections to replace NPO/xplot84driver within the interface. At present this is at an experimental stage and is not readily usable. It was suggested that an alternative approach would be to write a better interface for the existing NPO program. Overall the response was positive for the new program to be developed further, with useful code being taken from NPO. 6. Use of log summary files in GUI Martyn reported that Sheila Peters (new DL programmer) had added HTML tags into log files for some programs using Kevin Cowtan's libhtml libraries, and was now adding tags to identify information in logfiles which can be extracted e.g. by the interface to provide a summary. The tags are in the form of HTML comments so they are invisible to netscape etc. It was pointed out that this was similar to the old xloggraph "$$summary" tags. The problem remains of which information should go into the summaries. Martyn's suggestion was to start with the minimum amount of summary information and use feedback from users to add to it in future. Alun pointed out that there is no NT equivalent of the unix "grep" command, although there is one under Perl, and a suggestion was made for CCP4 should distribute a version of "grep". 7. Progress with REFMAC Garib was unable to attend due to illness so this item was dropped. 8. MOLREP, MAKECIF, LIBCHECK Alexei Vagin talked about his automated molecular replacement program MOLREP, which is 3/4 years old. It has some unusual features (for example, input of a priori knowledge of similarity and completeness of model) and a new function (spherically averaged phased translation function). Alexei also talked about the new monomer dictionary which will replace protin.dic - this will be a CIF-based dictionary, with three new programs to access and manipulate the data: LIBCHECK (to add new dictionary entries), SMILE2DICT (SMILES is a small molecule syntax for molecules, and SMILE2DICT creates dictionary entries from this) and MAKECIF (for setting up restraints). 9. Plans for next release The next release (3.6) is planned for sometime in November. The main additions are the introduction of Data Harvesting, which uses the new libharvest and libccif libraries. Other developments: new programs MOLREP (from Alexei), OASIS (MIR with direct methods) and SC (shape complementarity program from Mike Lawrence), plus new versions of AMORE (this will require changes to the interface), ARP/wARP, SFCHECK and RASMOL. CCP4i will also officially be part of the suite from this release. Longer term projects include: re_position (program from Kim Henrick for making biologically sensible complexes after molecular replacement), ROTGEN, LSCALE and LAUE, and Chris Gilmore's maximum entropy program EM (formerly MICE). Some concerns were voiced about possible year 2000 compliancy issues, especially with regard to the Data Harvesting and the harvesting audit date. It was suggested that date format issues might also be a problem at the deposition stage. 10. Date of next meeting January 26th 2000 11. AOB Automation of the heavy-atom search was mentioned by Eleanor, and it was mentioned that Paul Emsley had written a program (CHART) with this in mind, however no-one present had tried to use it. It was suggested that since CCP4 already contains programs which are used to do heavy atom searches, maybe this is a possible future direction.