"Revolutionizing Protein Engineering with Computational Techniques and AI Integration"

MIT researchers have developed a computational approach to predict mutations that will lead to improved proteins, making it easier to engineer proteins with useful functions. Using a convolutional neural network trained on experimental data, the researchers created fitness landscapes to predict optimized versions of green fluorescent protein (GFP) and a protein from adeno-associated virus (AAV). The approach has the potential to accelerate the process of optimizing proteins for research and medical applications, and could be applied to other protein engineering problems.