Scholarlink

<p>The hepatitis B virus (HBV) infection continues to pose its own share of health challenges in the world with almost 296 million infected and over 820,000 liver related deaths happening every year. The existing drugs such as nucleoside analogues and interferon-alpha can suppress the replication but cannot eliminate the covalently closed circular DNA (cccDNA) reservoir, which requires a lifetime treatment and increases the risks of resistance and hepatocellular carcinoma (HCC). Drug repurposing is an alternative strategy, as it uses already approved drugs with a known safety profile to make therapeutic innovation faster. Scopus, Web of Science, PubMed, and Google Scholar (search terms: “HBV drug repurposing AI/ML,” machine learning HBV screening) were used as sources of the relevant articles. The review explains why artificial intelligence (AI) and machine learning (ML) play a central role in this field, discussing the basic strategies including supervised learning, deep neural networks, natural language processing, and reinforcement learning. Such methodologies enable high throughput virtual screening, predicting interventions between drugs and their targets, and stratifying patients to enable personalised interventions. Using examples, such as hybrid knowledge graph models to predict cures and graph neural networks on multitarget antiretrovirals in HBV-HIV coinfection, the discussion notes such progress as 85% functional cure forecasting accuracy and 70% hits on repurposed candidate identification. The obstacles, such as the biases in the data and regulatory barriers, are discussed and the prospects of federated learning and quantum-enhanced simulations are mentioned.AI/ML frameworks integrate computational pipelines with clinical validation processes, thereby accelerating development timelines by approximately 50–60% while promoting equitable access in resource-limited settings. This convergence enables the establishment of optimized therapeutic regimens and facilitates more effective and inclusive control of hepatitis B virus (HBV) infection.</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

Harnessing artificial intelligence and machine learning for accelerating drug repurposing in hepatitis B virus therapy

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月19日

查看全文
  • Samuel Chima Ugbaja,
  • Reuben Samson Dangana,
  • Idris Olatunji Sanusi,
  • Olukayode Adebola Ibitoye,
  • Asekho Nkungu,
  • Lwandiswa Mafuleka,
  • Moses Okpeku,
  • Nceba Gqaleni

摘要

The hepatitis B virus (HBV) infection continues to pose its own share of health challenges in the world with almost 296 million infected and over 820,000 liver related deaths happening every year. The existing drugs such as nucleoside analogues and interferon-alpha can suppress the replication but cannot eliminate the covalently closed circular DNA (cccDNA) reservoir, which requires a lifetime treatment and increases the risks of resistance and hepatocellular carcinoma (HCC). Drug repurposing is an alternative strategy, as it uses already approved drugs with a known safety profile to make therapeutic innovation faster. Scopus, Web of Science, PubMed, and Google Scholar (search terms: “HBV drug repurposing AI/ML,” machine learning HBV screening) were used as sources of the relevant articles. The review explains why artificial intelligence (AI) and machine learning (ML) play a central role in this field, discussing the basic strategies including supervised learning, deep neural networks, natural language processing, and reinforcement learning. Such methodologies enable high throughput virtual screening, predicting interventions between drugs and their targets, and stratifying patients to enable personalised interventions. Using examples, such as hybrid knowledge graph models to predict cures and graph neural networks on multitarget antiretrovirals in HBV-HIV coinfection, the discussion notes such progress as 85% functional cure forecasting accuracy and 70% hits on repurposed candidate identification. The obstacles, such as the biases in the data and regulatory barriers, are discussed and the prospects of federated learning and quantum-enhanced simulations are mentioned.AI/ML frameworks integrate computational pipelines with clinical validation processes, thereby accelerating development timelines by approximately 50–60% while promoting equitable access in resource-limited settings. This convergence enables the establishment of optimized therapeutic regimens and facilitates more effective and inclusive control of hepatitis B virus (HBV) infection.