Yadalam Pradeep Kumar, Ardila Carlos M
Department of Periodontics, Saveetha Dental College, Saveetha Institute of Medical and technology sciences, SIMATS, Saveetha. University, Chennai, Tamil Nadu, India.
Department of Basic Sciences, Biomedical Stomatology Research Group, Faculty of Dentistry, Universidad de Antioquia U de A, Medellín, Colombia.
J Oral Biol Craniofac Res. 2025 May-Jun;15(3):445-453. doi: 10.1016/j.jobcr.2025.02.012. Epub 2025 Mar 10.
Non-coding RNAs, including lncRNAs, circRNAs, and microRNAs, constitute 98 % of the human transcriptome and are vital regulators of gene expression, cellular processes, and host-pathogen interactions, particularly in viral infections. This study explores lncRNA-circRNA interactions and their biological significance in oral viral infections.
ViRBase, a database with over 820,000 interactions involving 50,000 RNAs from 116 viruses and 36 host organisms, was used to analyze herpesvirus datasets. The study employed hierarchical attention and knowledge graph embeddings to represent nodes and edges in the knowledge graph. These served as input features for a hierarchical attention model trained over 100 epochs. Model performance was evaluated based on loss calculation, optimization, and attention weight stability.
The model achieved a final loss of 0.000180 at Epoch 100, with stable attention weights confirming reliability. Node embedding statistics showed a mean of 0.005110 and a standard deviation of 0.013370, while attention weights had a high mean of 0.997178, emphasizing model robustness.
This study provides insights into lncRNA-circRNA interactions in herpes viral infections, enhancing therapeutic development, disease progression monitoring, and understanding host-pathogen interactions, paving the way for targeted interventions and improved outcomes.
非编码RNA，包括长链非编码RNA（lncRNA）、环状RNA（circRNA）和微小RNA，构成了人类转录组的98%，是基因表达、细胞过程和宿主-病原体相互作用的重要调节因子，尤其是在病毒感染中。本研究探讨lncRNA与circRNA的相互作用及其在口腔病毒感染中的生物学意义。
使用ViRBase数据库（一个包含来自116种病毒和36种宿主生物体的50,000种RNA的820,000多个相互作用的数据库）分析疱疹病毒数据集。该研究采用分层注意力和知识图谱嵌入来表示知识图谱中的节点和边。这些作为在100个轮次上训练的分层注意力模型的输入特征。基于损失计算、优化和注意力权重稳定性评估模型性能。
该模型在第100轮时最终损失为0.000180，注意力权重稳定证实了其可靠性。节点嵌入统计显示均值为0.005110，标准差为0.013370，而注意力权重的均值较高，为0.997178，强调了模型的稳健性。
本研究为疱疹病毒感染中lncRNA与circRNA的相互作用提供了见解，有助于促进治疗开发、疾病进展监测以及理解宿主-病原体相互作用，为靶向干预和改善治疗效果铺平道路。
