Kim, Seeun Oh, Simaek Woo, Hyeonuk Sim, Jiho Seok, Chaok Park, Hahnbeom 2025-08-20T06:36:37Z 2025-08-20T06:36:37Z 2025-08-20 2025-07 1758-2946 https://pubs.kist.re.kr/handle/201004/152992 Interactions of proteins with other molecules are often mediated by a set of critical binding motifs on their surfaces. Most traditional binder designs relied on motifs borrowed from known binder molecules, which highly restricted their applicability to novel targets or new binding sites. This work presents a deep learning network MotifGen that predicts potential binder motifs directly from receptor structures without further supporting information. MotifGen generates motif profiles at the receptor surface for 14 types of functional groups or 6 chemical interaction classes. These profiles are highly human-interpretable and can be further utilized as pre-trained embedding inputs for versatile few-shot binder design applications. We demonstrate MotifGen's effectiveness through its applications to peptide binder design and small molecule binding site prediction, where it either surpassed existing methods or added significant value when integrated. Our motif-centric approach can offer a new design strategy for novel binder discovery for challenging receptor targets. English BMC Deep learning molecular interaction motifs from receptor structures alone Article 10.1186/s13321-025-01055-8 1 Journal of Cheminformatics, v.17, no.1 Journal of Cheminformatics 17 1 Y scie scopus 001539975600001 2-s2.0-105012167600 Chemistry, Multidisciplinary Computer Science, Information Systems Computer Science, Interdisciplinary Applications Chemistry Computer Science Article PROTEIN-STRUCTURE DNA INTERACTIONS PREDICTION DESIGN ALGORITHM ATOMS Structure-based drug design Binding motif Peptide design Protein structure analysis