Adaptive multi-view learning method for enhanced drug repurposing using chemical-induced transcriptional profiles, knowledge graphs, and large language models

Journal of Pharmaceutical Analysis, Год журнала: 2025, Номер unknown, С. 101275 - 101275

Опубликована: Март 1, 2025

Язык: Английский

circ2LO: Identification of CircRNA Based on the LucaOne Large Model

Genes, Год журнала: 2025, Номер 16(4), С. 413 - 413

Опубликована: Март 31, 2025

Circular RNA is a type of noncoding with special covalent bond structure. As an endogenous in animals and plants, it formed through splicing. The 5′ 3′ ends the exons form circular at back-splicing sites. plays important regulatory role diseases by interacting associated miRNAs. Accurate identification can enrich data on provide new ideas for drug development. At present, mainstream recognition algorithms are divided into two categories: those based sequence position information biometric information. Herein, we propose method RNA, called circ2LO, which utilizes LucaOne large model feature embedding splicing sites sequences as well their upstream downstream to prevent semantic loss caused traditional one-hot encoding method. Subsequently, employs convolutional layer extract features self-attention mechanism interactive accurately capture core Finally, uses fully connected identify RNA. accuracy circ2LO human dataset reached 95.47%, higher than values shown existing methods. It also achieved accuracies 97.04% 72.04% Arabidopsis mouse datasets, respectively, demonstrating good robustness. Through rigorous validation, has proven its high-precision capability RNAs, marking potentially transformative analytical platform circRNA research field.

Язык: Английский

Computational modeling approaches and regulatory pathways for drug combinations

Drug Discovery Today, Год журнала: 2025, Номер unknown, С. 104345 - 104345

Опубликована: Март 1, 2025

Язык: Английский