Whither Structural Biology? - Gerard Kleywegt
On 22 July 2021, the existence of the AlphaFold DB was revealed in a publication (and a posting on the CCP4 bulletin board, of course) and the resource itself became freely and openly available on the same day (1).
From the start, it contained ~365,000 3D structures from over 20 near-complete proteomes, reliably predicted using the revolutionary AlphaFold2 method (2).
This development made some people question their study or career choices, but I will argue that in fact it may well herald a whole new Golden Age for Structural Biology (and structural biologists). There are a number of reasons for this, including the fact that there are still limitations to the prediction methods and that there is a need to validate the predictions (including using structure determination). Moreover, the availability of (soon over 100,000,000) models will provide many new opportunities in structural biology and beyond.
I will discuss these issues briefly and hope to convey a message of optimism to the next generation of structural biologists - your work methods and practices may change, but your productivity and impact may very well increase. In addition, there are many new challenges waiting to be tackled.
(1) Tunyasuvunakool et al., Highly accurate protein structure prediction for the human proteome. Nature 597, 590-596 (2021).
DOI:10.1038/s41586-021-03828-1. PMID: 34293799.
(2) Varadi et al., AlphaFold Protein Structure Database: massively expanding the structural coverage of protein-sequence space with high-accuracy models. Nucleic Acids Res., in press.
DOI:10.1093/nar/gkab1061. PMID: 34791371.