To predict 3D structure of proteins It is a long process to obtain high-quality structural data, leading to low quality results Researchers are yet to crystallize many biologically important proteins, due to - size, - solubility (for example, membrane proteins), - inability to express and purify in sufficient amount ### AlphaFold by Google's Deep Mind: AI-driven protein folding Challenges of classical computing-based simulation: - formation of protein complexes, - protein-protein interactions - protein-ligand interactions Quantum Compute allows for the explicit treatment of electrons enabling us to overcome the challenges.. the interactions are most difficult to classically solve Google’s AI model—which is trained on around 170,000 different structures of protein data—requires more than 120 high-end computers for several weeks.