Translated data: Recently, scientists published a study on the prediction of protein-ligand complex structures in the journal "Nature Machine Intelligence." The new method, NeuralPLexer, utilizes deep generative models to directly predict structures with only protein sequences and ligand molecular graphs as inputs. This method holds significant potential applications and can play a crucial role in the field of drug discovery. Through this research, an important step has been taken in predicting the structures of protein-ligand complexes, providing new possibilities for future pharmaceutical research and bioengineering.