{"ok":true,"nota":{"id":"fb1bc7dea2315855","slug":"una-inteligencia-artificial-propone-moleculas-contra-una-bacteria-resi-fb1bc7","titulo":"Una inteligencia artificial propone moléculas contra una bacteria resistente a casi todo","resumen":"Un equipo entrenó redes neuronales de grafos para predecir compuestos antibacterianos contra Acinetobacter baumannii. El modelo funciona casi tan bien como uno con descriptores artesanales y muestra qué átomos y conexiones pesan en cada predicción.","fuente":"biorxiv","fuenteNombre":"bioRxiv · preprints","color":"#be2628","autor":"Romano, S., Grassia, M., Righelli, D., Lio, P., Mangioni, G., Taccioli, C.","url":"https://www.biorxiv.org/content/10.64898/2026.10.26.761825v1?rss=1","imagen":"/hoy/portada/fb1bc7dea2315855","tema":"ciencia","subtemas":["resistencia antibiótica","inteligencia artificial","antibióticos","bacterias hospitalarias","descubrimiento de fármacos"],"entidades":["bioRxiv","Acinetobacter baumannii","Organización Mundial de la Salud"],"formato":"dato","entretenimiento":false,"pubTs":"2026-10-29T00:00:00.000Z","indexTs":"2026-10-30T00:12:24.320Z","hash":"0425835face424a8453adecdb8efe2a8a3af7f1c83a98788775372c58e3a6de9","tsa":{"time":"Oct 30 00:13:11 2026 GMT","provider":"freetsa.org","sobre":"sello"},"stats":{"views":0,"likes":0,"clicks":0,"dwell":0},"tipo":"motor","registro":"hallazgo","porQueImporta":"Acinetobacter baumannii es una bacteria hospitalaria que ya no responde a muchos antibióticos, y la OMS la marca como prioridad crítica. Que el modelo explique qué partes de una molécula le importan permite a los químicos decidir qué probar en el laboratorio, en lugar de confiar en una caja negra. Eso acorta el camino hacia un tratamiento que hoy hace falta en cualquier terapia intensiva.","quienFirma":"bioRxiv (preprint, equipo de aprendizaje automático aplicado a química medicinal)","confianza":"preprint","consulta":{"pregunta":"¿Confiarías en una IA para elegir qué remedio te den?"}},"sello":{"hash":"0425835face424a8453adecdb8efe2a8a3af7f1c83a98788775372c58e3a6de9","sello":"7f61d60726b54762eca122d2f7bf7cb0f9f324702ee76f2fc2fa27c6118751f4","beacon":{"drand":{"cadena":"8990e7a9aaed2ffed73dbd7092123d6f289930540d7651336225dc172e51b2ce","ronda":6596271,"aleatorio":"475c20d708c2524a0cdebe4af77dfec4e122e8884ff50a0b62e90561b365834d","tiempo":"2026-10-30T00:12:30.000Z"},"nist":{"cadena":2,"pulso":2005069,"salida":"B55044227417F6923433CBC2983B4E344B47070946A68BA8DC65AF11249229CBE5D9905DB414536D56361BCB8273863C09AB5DD19B3BEC1B620F044B5C5E59D8","tiempo":"2026-10-30T00:11:00.000Z"},"noAntes":"2026-10-30T00:12:30.000Z"},"verificacion":{"drand":true,"nist":true},"tsa":{"time":"Oct 30 00:13:11 2026 GMT","provider":"freetsa.org","sobre":"sello"},"original":{"titulo":"Interpreting Graph Neural Networks for Antibacterial Drug Repurposing against Acinetobacter baumannii","descripcion":"Antimicrobial resistance in Acinetobacter baumannii, whose carbapenem-resistant strains are classified by the World Health Organization as a critical priority, has motivated machine-learning screens for new antibacterials, but translating predictions into chemical hypotheses requires evaluating the molecular features on which a model relies. A graph neural network ensemble achieves competitive predictive performance against a descriptor-augmented reference model on the same antibacterial screen while using only atom features and connectivity, allowing interpretation to focus on the molecular graph without a separate input of handcrafted descriptors. The atom-level explanations identify featu","pubDate":"2026-10-29","url":"https://www.biorxiv.org/content/10.64898/2026.10.26.761825v1?rss=1"},"indexado":"2026-10-30T00:12:24.320Z"}}