Minutes of the Working Group 2 Meeting -------------------------------------- Wednesday Jan 21st 1998, York Chemistry Dept --------------------------------------------- Present: Phil Evans (Chairman) MRC, Cambridge Martyn Winn Daresbury/CCP4 Christine Cardin Reading Sue Bailey (CCP4 Secretary) Daresbury/CCP4 Fred Vellieux IBS/Grenoble Garib Murshudov York/CCP4 Alexei Vagin York Alun Ashton Daresbury/CCP4 Eleanor Dodson York Jim Raftery Manchester Liz Potterton York/CCP4 Sheila Gover Oxford Chris Gilmore Glasgow Paul Emsley Glasgow Tadeusz Skarzynski Glaxo Joe Jaeger Leeds Misha Isupov Exeter David Waller Sheffield John Barker Bristol Apologies: Peter Moody, Peter Keller, Lindsay Sawyer, Steve Yewdall, Paul Holden, Ian Tickle, Cameron Dunn, Jim Naismith,. 1) Minutes of meeting of 24-9-97 and matters arising ---------------------------------------------------- The minutes of the meeting were accepted. There were a number of matters arising which were discussed: 1. Progress on getting old proceedings on the web. 2. SGML/XML. 3. Publication of proceedings in Acta D, and distribution of these. 4. Newsletters and need to encourage contributions. 5. Archival of CCP4BB messages - a volunteer was sought but not found. 6. Importance of WG2 members reporting back to groups. 2) Study Weekend '99 --------------------- Venue - Chester? Topic - Data collection and processing Organisers - Leo Brady + Johann Turkenburg (but only if he can attend the St. Andrews WG2 meeting) After discussion of the venue it was suggested that Eleanor book York for the year 2000. ACTION Eleanor Some suggested topics and speakers were: Basic introductory talks - why need good data collection. Strategy of data collection (making the most of your beam time/preparing for your synchrotron trip) - Peter Kuhn? (http://biosg1.slac.stanford.edu/) influence of new detectors cryo-techniques (Elspeth Garman?) Martin Walsh (Argonne) - testing new detectors Z. Otwinowski - definitely should be asked Handling Crystals (eg cutting crystals) Dominique Bourgeois (IBS,ESRF) paper in Acta or JAC on use of Baye's for data processing Andy Thompson (IP and CCD detectors at ESRF) ACTION: Leo 3) Results of Study Weekend Questionnaire ----------------------------------------- Sue drew the committee's attention to the results of the informal questionnaire circulated at Reading. The results are intended to help us judge usage of certain programs and to give us a feel for the hardware, operating systems and software that our users are using. For example, widely used unsupported programs such as AREAIMOL, BAVERAGE and BPLOT would be moved to the supported category. It was also suggested that EXTEND should be moved to the unsupported category and users should be encouraged to use MAPMASK instead - although Eleanor thought MAPMASK might not work for very large maps. It was suggested that dynamic memory allocation be incorporated into MAPMASK and REFMAC. (BUT IS IT ALREADY IN MAPMASK?) The function of PSAVER may not be necessary as NCSMASK does the same thing. A general examination of the programs in these categories led to the suggestion that RWCONTENTS should be incorporated into PDBSET, PRMAP (renamed MAPDUMP) should become supported and ACT, CONTACT and DISTANG could be combined as they do virtually the same thing. ZEROED is part of the old solvent flattening procedure and is no longer needed. RESLICE has been replaced by MAPMASK. 4) CCP4 workshops. ------------------ a. MAD workshop at DL. This will be held in June/July. Date to be fixed in the near future. Programme organisers are Keith Wilson and John Helliwell with Colin Nave advising as he has attended the recent ESRF MAD workshops. Other Daresbury staff will be helping. Two days beam time on 9.5 will be used for measuring EXAFS spectra of crystals. It is not intended that users bring their own crystals. The budget for the meeting is 10-11K pounds b. Sue reported that WG1 have agreed that CCP4 should support the Bath summer school at a level of 5K pounds. Sue noted that Garry Taylor may also looking for people to help with the lecturing as well as financial support. c. Eleanor said that Gerard K. had also suggested a workshop for programmers to help them use CCP4 routines etc. > Subject: Suggestion from Gerard K > - one thing i forgot on my ccp4 questionnaire: > could ccp4 organise a short (2 days ?) workshop > for prospective programmers to explain how > the subroutine libraries work (how to handle > mtz and map files; how to set up an fft; etc.); > i know this is all sort of documented but it > is much better to get some face-to-face > instructions by fellow programmers; if cost > is low (or free), you could attract young people > from all over europe who are in principle > interested in developing software for ccp4, > and every new program that uses ccp4 standards > (mtz, maps, etc) is good for ccp4 (even if the > program is not distributed through ccp4) This was agreed to be a good idea and CCP4 would be happy to support it. Ideally for 15-20 people, interested or preferably already involved in software development and not restricted to UK participants. The best way would be if Gerard would like to organise such a meeting in Uppsala. Eleanor suggested June to tag onto an existing Uppsala meeting. ACTION: Eleanor d. It was also noted that CCP4 is supporting John Helliwell's meeting celebrating 50 years of the computer. CCP4 have donated 1K pounds for Alwyn Jones to give a CCP4 lecture and to help fund students to attend. 5) CCP4 supporting an IUCR microsymposium ------------------------------------------ CCP4 has agreed to support a microsymposium at the IUCR 1999 Glasgow meeting. The available microsymposia topics will be decided in July, and Chris agreed to suggest a topic for CCP4 at this time, in consultation with Neil Isaacs. CCP4 were also keen to help support young scientists attending the computing school, probably by paying for travel from Glasgow and registration costs. Details will be decided later. Chris also asked what the level of attendance was likely to be for a microgravity satellite meeting or microsymposium. The members were not very enthusiatic. 6) VMS support -------------- With the departure of Adam, none of the Daresbury staff felt themselves fully competent to check and test the suite on a VMS system. A volunteer was requested to do this. Sheila said that Richard Bryan at Oxford would like to help but would like to know more of what was required. Martyn said it would involve installing new versions of the suite prior to official release and feeding back problems (preferably with solutions). Sheila wanted to know if CCP4 would provide any financial support for this work, but it was noted that CCP4 is supposed to be a collaborative project with University groups contributing to the effort and that Unix sites (eg York and Cambridge) which have a similar, but more significant, function do not receive financial support. Sheila agreed to take this information back to Richard. ACTION: Sheila The majority of the members felt that supporting Linux was likely to be much more important for the next few years and porting the suite to windows NT systems might be needed at some stage in the future. 7) Anomalous differences ------------------------ Eleanor explained that CCP4 uses anomalous differences in the form meanF, deltaF whereas most other software uses F+ and F-. Truncate will already output F+ and F- and Eleanor suggested changing other programs to use F+ and F- as well as the existing format. It was agreed that programs should have both options initially. Programs which require changing are SCALEIT, FHSCAL, MLPHARE. FFT might require some changes. Phil agreed to ask Eric to make a similar change to SHARP so that it is compatible. It was noted that people should always run TRUNCATE with option anomalous on. ACTION: Eleanor 8) Efficient Numerical Routines. -------------------------------- Martyn said that he, Garib, Ian and Kevin all had requirements for using numerical routines and were using routines obtained from NETLIB (BLAS, LAPACK, ARPACK and FFTW). (http://www.netlib.org/liblist.html). It was proposed that these routines be used as a standard in CCP4 programs. Generic forms of the routines could be freely distributed with CCP4, and machine specific forms (faster) were available on many machines. The routines are thought to be better optimised and more robust than routines we currently use and could replace much of MODLIB. It was agreed that these routines would be distributed with CCP4 and that existing programs could be changed to use them. ACTION: Martyn 9) New programs for V3.4 ------------------------ Martyn described the new programs that we expect to include in V3.4. These are RANTAN (direct methods program from Yao), SFCHECK (from Alexei Vagin), ROTGEN (an X-windows program for generating X-ray diffraction rotation images from John Campbell) and FALLOFF (for visualising anisotropy of data from Yorgo Modis). It was felt that, with the permission of the authors, FALLOFF would be better incorporated into TRUNCATE rather than as a stand-alone program. 10) Progress on Maximum Entropy work ------------------------------------ Chris described recent progress on the Maximum Entropy work. The program has been converted to the macromolecular case, and runs on a variety of unix platforms. A Tcl/Tk interface has been prepared and Chris will talk to Liz about integrating it into the CCP4 GUI. Testing has just begun on real macromolecular examples. 11) GUI update -------------- Liz distributed a list of programs for which interfaces will be available in the initial release of the GUI, and most of these have already been done. A beta-release is planned in 1-2 months, and 4-5 sites have been volunteered to help in this. A utility "loggraph" from Darren Spruce, which uses the blt extension to the standard Tcl/Tk, will be used to display 2-D graphs, etc. The latest Rasmol with map drawing capabilities is now the responsibility of Mike Hartshorn. Tadeusz said that Mike was willing to work on Rasmol, but needed someone to tell him what was required. Molview was also to be investigated as a possible 3-D viewer. 12) Progress on new coordinate format ------------------------------------- Martyn and Liz have been detailed to work on the new mmCIF-like format, but due to other committments have not made much progress. They are working on an intermediate level library that will interface between application programs and Peter Keller's low-level mmCIF routines. A meeting of the coordinate format sub-committe is planned for February/March. Phil and Eleanor mentioned several problems with the new rwbrook routines released with 3.3. These will be patched, but no major re-write is planned due to the imminent change of format. ACTION: Martyn, Liz 13) Date of next meeting ----------------------- This will be at the BCA meeting St Andrews on Tuesday 7th April, 9.00am - 1.00pm. 14) Any Other Business. ---------------------- Phil mentioned that Neil Isaacs wanted a "What is CCP4 for?" meeting to discuss the future direction of CCP4, and how the younger generation can be brought into the fold. There were many opinions on this, which were freely aired during the meeting, down in the pub, and on the train. =================== Sue Bailey Martyn Winn 22.1.98