Researchers from the University of Pennsylvania and collaborators introduced PepMLM, an AI model leveraging masked language modeling to design peptide binders using only protein sequence data, circumventing traditional dependency on protein structural information. Published in Nature Biotechnology, PepMLM successfully generated peptides that bind challenging therapeutic targets implicated in diseases such as Huntington’s, leukemia, and viral infections. Its widespread adoption reflects the model’s accessibility and potential to open new avenues in drug discovery targeting elusive protein interactions.