It has always been popular to start companies around families and subfamilies of targets, banking on the similarity among the receptors to speed drug discovery. But the idea largely hasn't panned out, in part because of the new target risk--sometimes they're not pharmaceutically relevant and sometimes they resist the available chemistries. That's why G-protein coupled receptors (GPCRs) have been so important: they're clearly relevant (many, perhaps most, blockbusters come from this class of receptor) and their unique structure makes them both relatively easy to hit with ligands and likely to do something when hit. But the same structural advantages turn into scientific disadvantages for researchers: they resist screening and other techniques of modern drug discovery. We explore some of the newest approaches to mining this rich vein of opportunity.

Start-ups bent on determining the 3D structures of proteins are trying to figure out the best ways to package their capabilities and data for sale to drugmakers. Growing numbers of commercial and academic researchers are seeking information about proteins--in particular structural information that influences how these important actors of the body function, and how drugs may fit them. X-ray crystallography is the gold-standard way of getting structural information, and a number of start-ups are gearing up to use the method on a larger scale than ever before. They want to sell the data to drugmakers to use in drug discovery, or better yet, form alliances to obtain and leverage the data-whose value in quantity is as yet unproven. Business models are still evolving: early plans to create databases are fading as firms acknowledge it will take time to accumulate quantities of structures and drugmakers already deluged with data demand actionable information. Service models may suffice temporarily, but won't provide much return in the long-term. Small start-ups are focusing on determining structures of known targets, or new but simple ones, while other better-capitalized companies lay bigger plans: encouraging drugmakers to look at unknown targets and learn more broadly, or to utilize multiple crystals. All the firms are getting involved in screening as a means of lead discovery. Companies starting up in the field of structure-based drug design are competing in the marketplace not only with each other, but against the availability of public data, and other crystal-free methods of gaining information about proteins.