Application of AlphaFold3: Innovative Protein Structure Prediction Toolin Research and Education

Artificial intelligence (AI) has a significant impact on our daily lives, including the field of chemistry and the molecular life sciences, with structural biology as a prominent example. The 2024 Nobel Prize in Chemistry recognizes the advances made through an AI model called AlphaFold2 (AF2). AF2 can predict protein 3D-structures with near-experimental accuracy, which was previously notoriously difficult. AF2 has been widely used to understand the structural and biological functions of proteins. Despite the wide applications in the life sciences, there were challenges with AF2, such as protein interactions with other small molecules. To overcome the limitations of AF2, the third version of AlphaFold, AF3, was developed. AF3 has the ability to predict protein structures as well as protein-protein, protein-nucleic acid, and protein- ligand interactions with high confidence. These capabilities are essential for understanding biological phenomena and the development of drugs targeting various diseases or disorders. Here, we briefly compare and contrast AF2 and AF3. The implications of the broad applications in structural biology and biochemistry education are discussed, highlighting AF3 in these fields. We present examples of applications of AF3 from two life sciences classes at a resource-limited institution, implying potential in undergraduate education. Finally, we address the limitations and challenges of AF3 applications and explore possible future directions.