13 February 2024 | Qiqige Wuyun, Yihan Chen, Yifeng Shen, Yang Cao, Gang Hu, Wei Cui, Jianzhao Gao, and Wei Zheng
The review provides a comprehensive overview of recent advancements in protein tertiary structure prediction, highlighting the integration of artificial intelligence (AI) algorithms, particularly deep learning methods. The article discusses various methodologies, including template-based modeling (TBM), template-free modeling (FM), contact/distance-guided methods, end-to-end folding methods, and protein language model (PLM)-based methods. It also covers multi-domain protein structure prediction, the Critical Assessment of Protein Structure Prediction (CASP) experiments, and the AlphaFold Protein Structure Database (AlphaFold DB). The review emphasizes the advantages and disadvantages of each method, aiming to guide future research in protein structure prediction. Key milestones, such as the introduction of AlphaFold2 and its performance in CASP14, are highlighted, along with the limitations and areas for improvement, particularly in multi-domain protein structure prediction. The article concludes by discussing the broader implications of these advancements for biomedical research and drug discovery.The review provides a comprehensive overview of recent advancements in protein tertiary structure prediction, highlighting the integration of artificial intelligence (AI) algorithms, particularly deep learning methods. The article discusses various methodologies, including template-based modeling (TBM), template-free modeling (FM), contact/distance-guided methods, end-to-end folding methods, and protein language model (PLM)-based methods. It also covers multi-domain protein structure prediction, the Critical Assessment of Protein Structure Prediction (CASP) experiments, and the AlphaFold Protein Structure Database (AlphaFold DB). The review emphasizes the advantages and disadvantages of each method, aiming to guide future research in protein structure prediction. Key milestones, such as the introduction of AlphaFold2 and its performance in CASP14, are highlighted, along with the limitations and areas for improvement, particularly in multi-domain protein structure prediction. The article concludes by discussing the broader implications of these advancements for biomedical research and drug discovery.