Changing times in macromolecular crystallography

[E. N. Baker]The present debate over the release of macromolecular structural data (see letter by Alex Wlodawer, this issue) highlights the way this field has evolved. It also raises ethical and practical issues that sometimes seem to be in conflict.

It has always been an accepted principle among crystallographers that their data should be made available at the time of publication; this validates the results and makes the data readily accessible to other scientists. Macromolecular crystallography posed new challenges. The sheer size of the data files precluded publication in journals. The establishment of the Protein Data Bank (PDB) was a crucial, far-sighted event, but deposition remained largely voluntary.

Ten years ago the IUCr provided vital leadership, through its Commission on Biological Macromolecules, by formulating a set of criteria that were intended to encourage deposition of macromolecular data. Included in these were a proviso that coordinates could be withheld from release, after deposition, for up to a year (and structure factors for 4 years). This gave crystallographers time to gain benefit from the years of investment that went into inherently very risky projects, while ensuring that the data would soon be fully available. At the same time the IUCr, and individual crystallographers, continued to press journals to make deposition a condition of publication.

What has changed? The data are now of unparalleled value to biologists. Structure-based drug design is but one application and all kinds of creative uses are now being explored. Exploitation of the information coming from genome projects can only accelerate the value of structural analysis. The public good therefore argues for immediate release. On the other hand, the pressures imposed by industrial collaborations, or by the short-term nature of most funding for what is still risky, long-term research, remain. Much depends on the attitudes of journals and funding agencies. Some journals still do not require deposition and many do not check that deposition has occurred. A change in policy by only some journals may be counter-productive, prompting researchers to publish elsewhere, perhaps in journals that do not require deposition al all.

I believe that the IUCr should again take the lead. In my view, the first priority is that all data should be deposited. As regards the time of release (immediately upon publication, or after a short "hold") I personally favour immediate release. For any IUCr recommendation to have force, however, it must have the support of crystallographers worldwide. It is urgent, therefore, that members of our community make their views known, to me or to Wolfram Saenger and other members of our Commission on Biological Macromolecules.

Edward N. Baker