Protein structure prediction in drug discovery

When the results of DeepMind’s AlphaFold2 at CASP were announced in 2020, the scientific world was so amazed by how effectively it performed that “it will change everything” became the motto for this revolution [1]. As a result, it should come as no surprise that “Protein Structure Prediction” was named Nature’s Method of the Year 2021. Structure-based drug discovery (SBDD) is the one area of biology and medicine that is most expected to benefit and excel as a result of the developments of AlphaFold2 and comparable tools, such as RoseTTAFold [2]. However, since the accuracy of the residues’ conformations at the active sites remains a key limitation in SBDD, as does the inability to guess which conformational state of a protein these tools will predict, it is still necessary to associate and integrate previous physically based models and expert-driven knowledge with new machine-learning approaches, as well as experimentally derived structural data. New approaches and tools, as well as developments in previously existing techniques for protein structure prediction and applications in immunology and virology therapeutic intervention targets, are described in this Special Issue.