The day started by a lecture from Brussels by Remy Loris, who explained the impressive variety of oligomerization forms of lectins. A dazzling number of different structures passed the eye. This was followed by the beautiful right-handed helical structure of endopolygalacturonase II, presented by Yovka van Santen from Groningen. The structure determination was a very nice example of the current power of automated methods. Christel Verboven in the next talk showed that automation isn't yet capable of solving everything, as she explained the problems encountered in the structure determination of Vitamin D binding protein. Ivo Ridder went into the details of his 1.15 Angstrom dehalogenase structure, showing nicely the things that are possible when you have so much data, and where some of the errors in the refinement programs reside. Ruben Declerq ended the morning session with an interesting exposition on the possibilities of hexitol nucleic acids, and their structural details.
The methods session in the afternoon was started off suitably by the grand old man of Dutch protein crystallography, Jan Drenth himself. He gave an interesting introduction into his recent work on physico chemical aspects of the nucleation of protein crystals. He stressed that crystallization from a phase separation is sub optimal, because the condition is unstable and the protein concentration is relatively low.
Maxim Kuil then presented the interesting plans in Leiden to build a miniature crystallisation robot, making use of nanolitre volumes. Compared with other array based nano-techniques such as DNA chips they are faced with the formidable problem of keeping everything in solution. They hope to have solved this problem by making use of inkjet technology, which sounds very promising, although no tests were done as yet. Foreign guest speaker Randy Read (now at Cambridge, UK), then presented some brand new ideas on the use of maximum likelihood for molecular replacement. This becomes more and more interesting as more structure are known and the possibilities of refinement from lousy models become better. In test calculations there is a real benefit in using the new methodology.
After tea Dominique Maes presented work on different TIM structures with varying thermostability. It is impressive how little change there is between heat and cold-labile proteins. Finally Martin Weik presented the work done in collaboration with the team in Grenoble on glass transitions in protein crystals, upon changing the freezing temperature. He also presented some of the spectacular radiation damage studies done at the ID14-4 beamline in Grenoble. These show clearly how disulfide bonds get lost and glutamate and aspartic acid residues lose their carboxylates.
Sponsorship by Nonius and Bruker made invitation of a foreign guest speaker, Randy Read, and a small party at the end of the day possible, which certainly contributed to the overall enjoyment of the day.
The next meeting will be held in May 2000 in Groningen, hosted by Bauke Dijkstra.