Protein family membership governs exosite predictability across the structural proteome.

Vista sencilla de ítem
dc.contributorFOLORUNSHO BRIGHT OMAGE, UNIVERSIDADE ESTADUAL DE CAMPINAS; IVAN MAZONI, CNPTIA; INACIO HENRIQUE YANO, CNPTIA; GORAN NESIC, CNPTIA.
dc.creatorOMAGE, F. B.
dc.creatorMAZONI, I.
dc.creatorYANO, I. H.
dc.creatorNESHICH, G.
dc.date2026-03-23T11:48:39Z
dc.date2026-03-23T11:48:39Z
dc.date2026-03-23
dc.date2026
dc.date.accessioned2026-07-07T04:18:37Z
dc.descriptionExosites, defined as protein surface regions that mediate macromolecular recognition at sites distinct from catalytic centers, represent emerging targets for selective drug design, yet their structural diversity has precluded systematic computational identification. Here we demonstrate that exosite prediction performance varies substantially across protein families, ranging from Matthews correlation coefficient (MCC) of 0.47 for coagulation factors to 0.14 for kinases. Using ExositeDB, we developed STINGExoFind, a gradient boosting framework leveraging 87 structural descriptors from the STINGRDB2 database, and evaluated 180 proteins under leave-one-protein-out cross-validation (LOPO-CV). Coagulation proteases achieved 50% success rates at the MCC ≥ 0.5 threshold, whereas kinases and caspases remained largely unpredictable. Ten structures spanning six families exceeded MCC ≥ 0.7, including MAPK/ERK2 (MCC = 0.86) within the otherwise challenging kinase family, indicating that high-confidence predictions remain achievable for specific proteins even in poorly-performing families. These results establish exosite prediction as a family-specific rather than universal challenge: computational approaches can meaningfully guide experimental validation for coagulation factors and similarly consistent protein families, while structurally diverse families require experimental characterization. STINGExoFind is provided as a community resource to support future method development and exosite-targeting drug discovery.
dc.identifierArtificial Intelligence in the Life Sciences, v. 9, 100166, June 2026.
dc.identifier2667-3185
dc.identifierhttp://www.alice.cnptia.embrapa.br/alice/handle/doc/1185675
dc.identifierhttps://doi.org/10.1016/j.ailsci.2026.100166
dc.identifier.urihttp://hdl.handle.net/123456789/457370
dc.languageeng
dc.rightsopenAccess
dc.subjectAprendizado de máquina
dc.subjectEstrutura proteica
dc.subjectAumento de gradiente
dc.subjectDescritores de nanoambiente
dc.subjectDescoberta de fármacos
dc.subjectExosite prediction
dc.subjectMachine learning
dc.subjectGradient boosting
dc.subjectNanoenvironment descriptors
dc.subjectDrug discovery
dc.subjectProtein structure
dc.titleProtein family membership governs exosite predictability across the structural proteome.
dc.typeArtigo de periódico

Archivos

Colecciones

Embrapa Agricultura Digital (CNPTIA)