\def\Ang{\thinspace\AA} \def\Angsq{\Ang$^2$} \def\percent{\thinspace\%} % \magnification = \magstep1 \font\bfbigten=cmbx10 scaled\magstep1 \font\bfbignine=cmbx9 scaled\magstep1 \font\itbigten=cmbxti10 scaled\magstep1 \font\scbig=cmbx7 scaled\magstep1 \font\sc=cmr7 \font\scl=cmr7 scaled\magstep1 \hoffset .46 true cm \voffset .46 true cm \hsize 16.0 true cm \vsize 23.0 true cm % % \baselineskip=12pt \normalbaselineskip=12pt % \outer\def\section#1\par{\vskip0pt plus.3\vsize\penalty-250 \vskip0pt plus-.3\vsize\bigskip\vskip\parskip \message{#1}\centerline{\bf#1}\nobreak\smallskip} % \outer\def\subsection #1 #2\par{\medskip\centerline{#1 \it#2}\nobreak\smallskip} % \outer\def\subsubsection #1 #2\par{\medskip\leftline{#1 \it#2}\nobreak\smallskip} %\raggedright \centerline{\bf MINUTES OF A MEETING OF WORKING GROUP 2 OF CCP4} \centerline{\bf WHICH WAS HELD AT BIRKBECK College} \centerline{\bf on the 11th July 1990} \bigskip \noindent Present: \medskip $$\vcenter{ \halign{#\hfil&\qquad#\hfil\cr David Rice (Chairman)&Sheffield University\cr Don Akrigg &Leeds University\cr Peter Brick &Imperial College\cr Peter Daly &Daresbury Laboratory\cr Eleanor Dodson &York University\cr Phil Evans &MRC, Cambridge\cr Andrew Lapthorn &Liverpool Polytechnic\cr Colin Nave &Daresbury Laboratory\cr Jim Raftery &Manchester University\cr Paul Taylor &University of Edinburgh\cr Lynn TenEyck &San Diego Super Computer Centre\cr Ian Tickle &Birkbeck College\cr Francesca Young &University of Glasgow\cr Wojciech Wolf &Berlin\cr Kim Henrick (Acting Secretary)&IRC-Cambridge\cr }}$$ \bigskip \bigskip {\it 1. The minutes of the meeting of 2nd May 1990 were agreed.} \medskip {\it 2. Matters Arising} \medskip None. \bigskip {\it 3. New CCP4 Secretary} \medskip The Chairman introduced Wojciech Wolf to the working group and welcomed his appointment. Wojciech Wolf will take over the position of secretary to CCP4 from the 1st of December of this year. He introduced himself to the group as a crystallographer with a background in phosphorous heterocyclic chemistry and was currently working in Berlin on protein crystallography on a Humbold Fellowship. \bigskip {\it 4. 1991 Study Weekend.} \medskip Phil Evans outlined the program for this meeting as essentially as that set out in the minutes of the meeting of the 2nd of May. He confirmed that Wayne Hendrickson would be a speaker and indicated that the final list of speakers had yet to be compiled and was open to suggestions, particularly for speakers from different laboratories, to present the diversity of techniques available. Eleanor Dodson presented the view that a chemistry orientated talk on general protein metal binding would be invaluable and suggested approaching Huber's laboratory. Keith Wilson was also put forward as a possible speaker, to describe the Hamburg experience. A broad discussion ensued covering possible topics and speakers, including Patterson search methods, error screening, use of centric data, anomalous scattering, scaling and absorption corrections. Phil Evans indicated that he would present a general survey of phase/refine methods and Ian Tickle could discuss direct space Patterson vector methods. Andy Thompson was suggested as a speaker, to give a short talk on the machine requirements and the actual experimental techniques involved in collecting Anomalous data. Possible speakers to present case studies were then discussed. Phil Evans indicated that he would welcome suggestions for speakers and he would use the opportunity of the Bischenburg and Bordeaux meetings to finalise the speakers list. \bigskip {\it 5. CCP4 funds} \medskip Peter Daly and Colin Nave briefly detailed the present state for CCP4 funds, and indicated that there were in excess of 20,000 pounds immediately available. This sum does not include funds held in suspense accounts by Shirly Lowndes, nor the annual running expenses of 8,000 pounds, which have yet to be given to CCP4. \bigskip {\it 6. Discussion on Ways to Utilise the CCP4 Funds} \medskip Eleanor Dodson made the suggestion that part of CCP4's income could be used for the printing of extra copies of all the past CCP4 Study Weekends, and distributing a number of copies to each of the U.K. groups. The need for reproducing the proceedings as a reference set, was evident from experience of the recent Oxford Summer School on protein crystallography, where there was repeated reference to these proceedings, but complaints that these volumes were difficult to access. The secretary expressed the view that the proceeedings were widely distributed to all U.K. groups and that copies should be available. The meeting agreed that this should be a priority for the new CCP4 secretary to investigate the costs and a timetable for such an undertaking with the appropriate Daresbury reprographics staff. There followed a vigorous discussion on ways of making use of CCP4 funds, by making payments to PhD students for work of general interest, covering standards, supervision and documentation. However, the debate was inconclusive, floundering over a satisfactory payment procedure. The Chairman then suggested that each group make a list of programs that were in each in each laboratory but not currently part of CCP4. This should be sent to Peter Daly at Daresbury, giving the program function and documentation. The programs should be classified under the Overview headings as set out in the CCP4 Newsletter No. 23. \medskip \hrule \smallskip \noindent {\bf Post Meeting Note:} \smallskip {\scl The acting secretary has had discussions with Daresbury Staff over the future of the CCP4BB and the CCP4 programs available on the Daresbury INFO-SERVER. It was pointed out that this system was running on an aging GEC system and the continued support to CCP4 (and others) would require conversion of the mail software to a unix system. It may be politic for CCP4 to make a contribution to this conversion work, perhaps by part payment for a disk to store both the vms and the unix versions of CCP4. } \smallskip \hrule \bigskip {\it 7. Group Reports} \medskip Phil Evans described a new version of the program PROTIN. This incorporates Eleanor Dodson's additions of non-bonded symmetry related contacts into the constraints list and her keywording, together with Andrew Leslie's incorporation of carbohydrate geometry constraints. He also reported on a new ``jiffy'' program for placing Patterson vector sites into the asymmetric unit (making use of MSYMOP). There were also changes to the his MRC-FRODO program, in the use made of centric lattices. Phil Evans also indicated that there were errors in the Daresbury distributed RWBROOK library file, and was critical of the distribution of changed key library routines before rigorous testing had been carried out. He also detailed a problem Andrew Leslie had encountered using Eleanor Dodson's SF-routines with the space group R32, in that the atomic temperature factors produced were unexpectedly high. Andrew Leslie recommended using the DERIV program. \medskip \hrule \medskip \noindent {\bf Post Meeting Note:} \smallskip {\scl The Daresbury programs are checked with a minimum of testing, with only one data set, under a set of options that the Secretary was familiar with. The error in the RWBROOK library was not evident under this test. This highlights the problem Daresbury has in the lack of test data for a wide range of program options. The SF-routines actually only produce an error when compiled under vms fortran with the ``standards'' qualifier (truncating at column 72), and clearly demonstrates that programs written in VAXTRAN cannot then be treated as fortran77. There is a toolpacked version of the library routines and the space group 152 specific program available, in which the temperature factor error does not occur.} \smallskip \hrule \bigskip Andrew Lapthorn related problems workers at Liverpool had encountered using the MRC-FRODO (version 4.3) in the use of certain options, and the corrupting of files. There followed a brief discussion on possible bugs and/or incorrect use of commands. Andrew also asked in general about Alwyn Jones' ``O'' program availability and licence details, however no U.K. groups were yet using this package. Fran Young reported on her continuing work in porting CCP4 programs to an IBM system and will be able to pass a list of the changes back to Daresbury. A problem with the program PLUTO was discussed, concerning a possible problem in plotting atoms onto a map, at edge boundaries. An appeal for other general contouring programs/routines was made. Colin Nave briefly announced that John Campbell (at Daresbury) had an XVIEW version of the programs within the Laue suite that require film image display. Eleanor Dodson expanded on her work on the program PROTIN, for the checks on symmetry equivalent non-bonded interactions. She also related a problem encountered with water molecules in the calculation of derivatives in PROLSQ. Peter Brick indicated that problems can be resolved by treating each water molecule as a separate residue. She also described the new version of SYMOP.LIB wherein primitive and non-primitive symmetry operators were differentiated and could be separately accessed by the calling program. Changes were also made to the FFT program to check that, where sensible only the primitive operators were used. Don Akrigg floated a possible problem met with the running of FRODO at Leeds, in write back on a serial line. The suggestion was made that this was a hard ware fault not software. He also indicated a problem in making PLUTO give correct red/green stereo output. This was however an option for PLUTO that few others had tried. Ian Tickle related a dissatisfaction with Toolpacked code, in that the subroutine order was now alphabetical and that variable names, in particular Do Loop counters, had been changed from the original code. He also found errors in the use of the vms version of CCP4 command PRD, in that directory names for documentation had been changed. A related problem was found in TIDYTEXT. The secretary answered by indicating that all programming effort at Daresbury was now carried out on unix machines, where the compiling of large single fortran source files is impossible, with various line limits within the one file of between 4,000 and 10,000 lines. Also the making of object libraries under unix that file names are indexed, NOT the names of routines within a file, therefore, libraries had to be made by compiling a separate file for each routine. Therefore use was made of the unix ``fsplit'' command, to separate a single source file into separate routines for editing and compilation, followed by concatenating the component subroutines back into the stored source. The changes to loop counters was made to make the code more understandable, as the secretary had difficulties in following a variable such as ``I'' over 10 screens of code. The secretary apologies for the errors propagated in the renaming of the documentation directories, to the command PRD. The command is one he had not only never used but didnt know existed. At Daresbury John Campbell has been making changes to TIDYTEXT to allow the program to work on both vms and unix, to all possible output devices. Users should appreciate the difficulties in sending a file that is fortran Carriagecontrol to a general postscript laser printer under unix, and the distributed vms version of CCP4 was probably made during this development. Ian went on to describe some of the work he was doing on molecular replacement with multi-subunit proteins and partial translation functions. Peter Daly reported that the unix/vms plot84, plot9x was now working at Daresbury. He also indicated that he was investigating the public domain graphics package VOGLE, which appeared very powerful. He related a problem with the LMB-HPPLOT routine. This was due apparently to the LMB plotter being connected as a serial line to one terminal, without making use of spooling. The secretary reported that there was now available an EDTPLUS lookalike editor running under unix. This involves a close approximation to the vms editor keybinding onto GNU-EMACS. This is running on the DEC5400 at Cambridge, a DEC3100 in Hamburg and Peter Daly had implemented it on the Convex, Iris' and Sun's at Daresbury. There followed a general disscussion on how best to keep Daresbury up to date with definitive versions of the CCP4 programs. There was an appreciation that this was a major problem, requiring feedback to Daresbury from both the CCP4 community and increasingly from outside the U.K. \bigskip {\it 8. Background to the NSF Initiative in Protein Crystallography Computing} \medskip Lynn TenEyck was invited to participate in the CCP4 working group and give a general background to the computing project he was involved with. He describe the National Science Foundation (USA) funded project in protein crystallography computing, involving the San Diego Super Computer Centre (SDSCC), The University of California at San Diego (Chemistry Department) and the Scripps Institute in San Diego. The project evolved from the observation that within the San Diego area, there were twelve crystallographic groups, each having little contact and duplicating software work. The project is aimed at integrating software, standardisation, documentation, to yield a complete protein crystallography package. The project is to draw on the graphics strengths of the contributing laboratories to produce new interactive graphics programs making use of the ANSI standard graphics meta file. Portability was stressed, incorporating widely available routines, such as the NCAR X-windows subroutine library. Lynn TenEyck expressed his dissatisfaction with the interactive graphics programs currently in use, such as FRODO, which still rely on an inefficient vector stroke mode of drawing. The NSF funded project will look into ways of taking advantage of the raster techniques for the visualisation of X-ray derived data, for example, transparent electron density maps. The SDSSC was as aware as U.K. workers in the lack of a full quad buffer stereo graphics system from manufacturers and were attempting to interest companies in designing true stereo systems, rather than inefficient add-ons to existing machines. Lynn TenEyck expressed interest in layering the code with a relational data base manager to screen reflection data as required. Currently such systems were expensive in disk usage and in commercial licences and had limited query structures. The SDSCC project was also involved in the setting up of workshops. The first of these was to be held in September on the problem of convergence during refinement. He declared a willingness to cooperate with both CCP4 and the ESF initiative in protein crystallography and offered facilities for the checking and standardisation of source code and for the production of a professionally designed documentation set for CCP4. The chairman expressed the working groups appreciation to Lynn TenEyck for his overview presentation. He cautioned though about a full cooperation with the USA groups as he was concerned with CCP4 retaining its identity as an SERC/U.K. group. There was no agreement reached at the meeting on defining areas of common interest or for cooperation or in defining file system standards as this would have to wait until the new CCP4 secretary took up his post. There followed a discussion on how to set up master versions of protein crystallographic programs accesable by electronic mail from Daresbury of both CCP4 and SDSCC programs. Lynn TenEyck offered assistance with unix revision control tools and in the testing of pre-release versions. For unix systems it is currently feasible to make the update procedure at end user sites entirely automatic via the use of ``crontab'' and e-mail. Such a system could easily interrogate Daresbury for new versions, extract files and compile/install the programs. This would require a strict checking procedure before release of new versions. The chairman expressed the obvious link between CCP4/ESF and the SDSCC in the new MTZ file system format and there was an agreement to continue discussions at the Bischenburg meeting with Keith Wilson. Don Akrigg inquired about the possibility of a ``px'' workstation with the programs running with X-windows. The secretary felt that this was still 2 to 3 years in preparation. \medskip A date for the next meeting was set for Wednesday the 26th of September 1990, at York University, Department of Chemistry for 11-30 am. \bigskip \medskip \centerline {K. Henrick} \centerline{(Acting - Secretary)} \bye