The structure of protein is determined by the sequence, and the function of protein is determined by its structure. The advent of accurate protein structure prediction tools has created new opportunities and challenges in the fields of structural biology, structural bioinformatics, drug discovery and many other fields of life sciences. The accuracy of single-chain protein structure prediction has reached a level comparable to that of experimental methods. In this review, we provide an overview of the theoretical basis, development history, and recent advances in the field of protein structure prediction. Additionally, we discuss how the large number of predicted protein structures and artificial intelligence-based methods affect experimental structural biology. Open questions and future research directions in the field of protein structure prediction are analyzed.