Scholarlink

<p>Japanese encephalitis virus (JEV), a single-stranded RNA virus, poses a serious health threat across Asia, particularly among children. It can cross the blood–brain barrier, causing severe neurological disorders, and affects over a billion people annually, resulting in thousands of deaths. Advances in drug discovery-such as drug repurposing, high-throughput screening, and de novo drug design-have identified promising therapeutic candidates. Quantitative structure–property relationship (QSPR) modeling has emerged as a valuable tool in drug design, enabling the prediction of physicochemical properties based on molecular structure. Topological descriptors (TDs), particularly degree-based descriptors, offer insights into the molecular structures and physicochemical properties of compounds, enabling the prediction of drug efficacy in Japanese encephalitis (JE) treatment. In this study, we evaluated six anti-JEV drugs using degree-based TDs computed via the edge partitioning technique. Linear regression models were developed to correlate these descriptors with experimentally determined physicochemical properties. To evaluate the correlation between TDs and physicochemical properties, centrality measures were also calculated. Our findings reveal strong correlations between TDs, centrality measures, and physicochemical parameters, demonstrating the predictive power of graph-theoretical approaches in rational drug design. Additionally, certain descriptors were shown to have higher prediction accuracy, providing a foundation for improving QSPR models customized for certain compounds. This study underscores the potential of integrating computational and mathematical chemistry in preclinical drug evaluation, streamlining the identification and optimization of effective anti-JEV therapeutics.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Estimation of physicochemical properties of Japanese encephalitis drugs via degree-based topological descriptors and regression models

  • 期刊论文

  • 发表日期：

    2026年7月18日

查看全文
  • Prerana Kalita,
  • Trishna Deka,
  • Khyati Goswami,
  • Eshika Purkayastha,
  • Venkata Satish Kumar Mattaparthi

摘要

Japanese encephalitis virus (JEV), a single-stranded RNA virus, poses a serious health threat across Asia, particularly among children. It can cross the blood–brain barrier, causing severe neurological disorders, and affects over a billion people annually, resulting in thousands of deaths. Advances in drug discovery-such as drug repurposing, high-throughput screening, and de novo drug design-have identified promising therapeutic candidates. Quantitative structure–property relationship (QSPR) modeling has emerged as a valuable tool in drug design, enabling the prediction of physicochemical properties based on molecular structure. Topological descriptors (TDs), particularly degree-based descriptors, offer insights into the molecular structures and physicochemical properties of compounds, enabling the prediction of drug efficacy in Japanese encephalitis (JE) treatment. In this study, we evaluated six anti-JEV drugs using degree-based TDs computed via the edge partitioning technique. Linear regression models were developed to correlate these descriptors with experimentally determined physicochemical properties. To evaluate the correlation between TDs and physicochemical properties, centrality measures were also calculated. Our findings reveal strong correlations between TDs, centrality measures, and physicochemical parameters, demonstrating the predictive power of graph-theoretical approaches in rational drug design. Additionally, certain descriptors were shown to have higher prediction accuracy, providing a foundation for improving QSPR models customized for certain compounds. This study underscores the potential of integrating computational and mathematical chemistry in preclinical drug evaluation, streamlining the identification and optimization of effective anti-JEV therapeutics.