Sir

Your News story “Protein structures hint at the shape of things to come” (Nature 435, 547; 200510.1038/435547a) projects an excessively optimistic view of structural genomics, which your readership might find misleading.

The claim that “structural genomics initiatives are ready to roll out protein structures in large numbers” should be treated with scepticism. An alternative view predicts that the number of protein structure determinations will peak and then decline, as the pool of easy targets (so-called ‘low-hanging fruit’) gets depleted.

Recent technological advances have considerably eased the transit through some of the bottlenecks of crystallographic analysis. However, the unique biochemical properties of individual proteins continue to present a formidable barrier to any automated procedure of structure determination. A better way to assess the outcome of high-throughput initiatives would be to consider their overall success rate, given as the ratio of structures solved to the number of targets attempted.

It is for the grant agencies to decide whether funding of structural genomics programmes constitutes a worthwhile investment. Perhaps the most serious point, though, is that the wider scientific community might come to consider all structural biology endeavours as straightforward. This is not the case, especially at the frontier of modern structural biology, where the crystallographic analysis of large macromolecular complexes still requires heroic efforts. In this respect, the image that accompanies your News story is of a protein that structural biologists would consider not particularly challenging, and that is simpler than the four protein structures reported elsewhere in the same issue, obtained through traditional structural-biology approaches.

When asked by King Ptolemy I about an easy way to learn geometry, the Greek mathematician Euclid famously replied that there was no royal road to geometry. Paraphrasing Euclid, we could affirm with equal confidence that there is no royal road to protein structure determination.