Scholarlink

<p>The past decade has witnessed an unprecedented convergence of exposomic technologies, population-scale genomics, and AI-enabled data science, creating the conditions for a new integrative discipline. Here we introduce ExposoGenomics, defined as the integrative study of how the genome and exposome, treated as jointly dynamic systems, interact across the life course to shape health and disease. ExposoGenomics moves beyond classical gene-environment interaction models by embedding high-dimensional, temporally resolved exposure data within a multi-omic and AI-enabled analytical architecture oriented toward causal discovery, mechanistic understanding, and translational application. We describe the conceptual foundations of this framework, its mechanistic architecture linking external exposures to genomic responses through physiologically based kinetic models and adverse outcome networks, and the analytical approaches, including causal machine learning, graph-based integration, and foundation models, required to realize its potential. We emphasize that computational prediction must be accompanied by rigorous empirical validation, and that findings must be grounded in biologically plausible, causally supported mechanisms. In conjunction with this Perspective, <i>Human Genomics</i> formally launches ExposoGenomics as a dedicated article category and invites submissions that advance this integrative agenda.</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

ExposoGenomics: integrating genome and exposome as jointly dynamic systems for causal discovery and precision health

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月21日

查看全文
  • Vasilis Vasiliou,
  • Nicholas Katsanis,
  • Giuseppe Novelli,
  • Juergen K. V. Reichardt,
  • Bassam R. Ali,
  • Maria Gazouli,
  • Sek Won Kong,
  • Hongyu Zhao,
  • Kiril Veselkov,
  • Bhramar Mukherjee,
  • Dimosthenis Sarigiannis

摘要

The past decade has witnessed an unprecedented convergence of exposomic technologies, population-scale genomics, and AI-enabled data science, creating the conditions for a new integrative discipline. Here we introduce ExposoGenomics, defined as the integrative study of how the genome and exposome, treated as jointly dynamic systems, interact across the life course to shape health and disease. ExposoGenomics moves beyond classical gene-environment interaction models by embedding high-dimensional, temporally resolved exposure data within a multi-omic and AI-enabled analytical architecture oriented toward causal discovery, mechanistic understanding, and translational application. We describe the conceptual foundations of this framework, its mechanistic architecture linking external exposures to genomic responses through physiologically based kinetic models and adverse outcome networks, and the analytical approaches, including causal machine learning, graph-based integration, and foundation models, required to realize its potential. We emphasize that computational prediction must be accompanied by rigorous empirical validation, and that findings must be grounded in biologically plausible, causally supported mechanisms. In conjunction with this Perspective, Human Genomics formally launches ExposoGenomics as a dedicated article category and invites submissions that advance this integrative agenda.