{"ok":true,"nota":{"id":"51d1afc7ab8c1d4b","slug":"un-modelo-de-ia-disena-anticuerpos-mirando-solo-las-secuencias-sin-est-51d1af","titulo":"Un modelo de IA diseña anticuerpos mirando solo las secuencias, sin estructuras 3D","resumen":"Desarrollaron LangAAI, un modelo de inteligencia artificial que diseña regiones de anticuerpos usando únicamente las secuencias de aminoácidos del anticuerpo y del antígeno. Lo compararon con enfoques que requieren la estructura tridimensional.","fuente":"biorxiv","fuenteNombre":"bioRxiv · preprints","color":"#be2628","autor":"Ashraf, F. B., Lonardi, S.","url":"https://www.biorxiv.org/content/10.64898/2026.10.19.760314v1?rss=1","imagen":"/hoy/portada/51d1afc7ab8c1d4b","tema":"tecnologia","subtemas":["inteligencia artificial","anticuerpos","modelos de lenguaje","biotecnologia","diseño de proteinas"],"entidades":["LangAAI"],"formato":"dato","entretenimiento":false,"pubTs":"2026-10-24T00:00:00.000Z","indexTs":"2026-10-24T15:51:15.058Z","hash":"592ac82055165f50161d0facc62aec93f452a763059cc4f5bc9f8f45b3ebc595","tsa":{"time":"Oct 24 15:51:35 2026 GMT","provider":"freetsa.org","sobre":"sello"},"stats":{"views":0,"likes":0,"clicks":0,"dwell":0},"tipo":"motor","registro":"hallazgo","porQueImporta":"Para un laboratorio argentino con recursos limitados, esto baja la barrera de entrada a la búsqueda de anticuerpos. Aun así, cada predicción sigue necesitando validación en el laboratorio antes de creerla.","quienFirma":"bioRxiv (preprint; institución principal no identificada en el resumen)","confianza":"preprint","consulta":{"pregunta":"¿Confiarías en un diagnóstico o tratamiento diseñado por IA?"}},"sello":{"hash":"592ac82055165f50161d0facc62aec93f452a763059cc4f5bc9f8f45b3ebc595","sello":"cfb83cc05d1e349f73c7fb1fdbe232fdfbec6c7a26d6138ca3d1f9867c2f3f39","beacon":{"drand":{"cadena":"8990e7a9aaed2ffed73dbd7092123d6f289930540d7651336225dc172e51b2ce","ronda":6580868,"aleatorio":"d446ab85d2a54a02312d75dedfa22ede70a463319299c8faa87805acc3ce8c52","tiempo":"2026-10-24T15:51:00.000Z"},"nist":{"cadena":2,"pulso":1997627,"salida":"93705C7BB1B7D67A616D78DAD026F17972E793E92107F852A41532A7BAD3299306AA3A88C81A5D59313E87EB39370ED775F0CF6DA33958D42DA2288B1F85A13B","tiempo":"2026-10-24T15:50:00.000Z"},"noAntes":"2026-10-24T15:51:00.000Z"},"verificacion":{"drand":true,"nist":true},"tsa":{"time":"Oct 24 15:51:35 2026 GMT","provider":"freetsa.org","sobre":"sello"},"original":{"titulo":"Sequence-only antigen context improves modeling of antibody hypervariable regions","descripcion":"The complementarity-determining regions (CDRs) of antibodies govern antigen specificity; consequently, CDR design should be performed conditional on the target antigen. Most published approaches depend on the availability of the antigen's three-dimensional (3D) structure, which is often difficult to obtain experimentally. In contrast, sequence-only models circumvent this requirement but are typically trained on antibodies without incorporating antigen context; exposure to antigen sequences alone does not necessarily entail learning the underlying interaction determinants. Here, we introduce LangAAI, a masked language model that conditions exclusively on antibody and antigen primary sequences","pubDate":"2026-10-24","url":"https://www.biorxiv.org/content/10.64898/2026.10.19.760314v1?rss=1"},"indexado":"2026-10-24T15:51:15.058Z"}}