Gero has introduced ProtoBind-Diff, a masked diffusion language model that generates novel drug-like molecules conditioned solely on protein amino acid sequences, foregoing the need for structural data. Trained on over one million protein-ligand pairs, the model demonstrated comparable or superior performance to structure-based methods on both well-characterized and low-data targets. ProtoBind-Diff's sequence-based approach broadens drug discovery capabilities to underexplored protein targets and supports collaborative development across oncology, immunology, infectious disease, and aging-related therapeutics.