This article details a novel computational method for designing sequence-specific DNA-binding proteins (DBPs), which are crucial tools in biotechnology and gene editing. The researchers developed a pipeline, incorporating tools like RIFdock and LigandMPNN, to generate small DBPs that can recognize arbitrary DNA sequences, a challenge unmet by previous computational redesign efforts. Experimental validation via yeast display and biolayer interferometry confirmed that the designed DBPs bind their target sequences with mid-to-high nanomolar affinity and high specificity, with a solved crystal structure closely matching the computational model. Furthermore, these designed DBPs demonstrated functionality in living cells, effectively acting as transcriptional repressors in E. coli and activators in mammalian cells, highlighting their potential as customizable tools for gene regulation. The study emphasizes the importance of sampling diverse protein scaffolds and precise geometric positioning for successful DBP design, suggesting their approach expands upon native binding solutions to create orthogonal, sequence-specific genetic regulators.

References:

• Glasscock C J, Pecoraro R J, McHugh R, et al. Computational design of sequence-specific DNA-binding proteins[J]. Nature Structural & Molecular Biology, 2025: 1-10.