Translating from Proteins to Ribonucleic Acids for Ligand-binding Site Detection.
Mol Inform
; 41(10): e2200059, 2022 10.
Article
en En
| MEDLINE
| ID: mdl-35577762
Identifying druggable ligand-binding sites on the surface of the macromolecular targets is an important process in structure-based drug discovery. Deep-learning models have been shown to successfully predict ligand-binding sites of proteins. As a step toward predicting binding sites in RNA and RNA-protein complexes, we employ three-dimensional convolutional neural networks. We introduce a dataset splitting approach to minimize structure-related bias in training data, and investigate the influence of protein-based neural network pre-training before fine-tuning on RNA structures. Models that were pre-trained on proteins considerably outperformed the models that were trained exclusively on RNA structures. Overall, 71 % of the known RNA binding sites were correctly located within 4â
Å of their true centres.
Palabras clave
Texto completo:
1
Colección:
01-internacional
Base de datos:
MEDLINE
Asunto principal:
Proteínas
/
Redes Neurales de la Computación
Tipo de estudio:
Diagnostic_studies
/
Prognostic_studies
Idioma:
En
Revista:
Mol Inform
Año:
2022
Tipo del documento:
Article
País de afiliación:
Suiza
Pais de publicación:
Alemania