Uzuner Odongo Dilara, İlgün Atılay, Bozkurt Fatma Betül, Çakır Tunahan
Department of Bioengineering, Gebze Technical University, Gebze, Kocaeli, Turkey.
Commun Biol. 2025 Mar 27;8(1):502. doi: 10.1038/s42003-025-07941-z.
Generating condition-specific metabolic models by mapping gene expression data to genome-scale metabolic models (GEMs) is a routine approach to elucidate disease mechanisms from a metabolic perspective. On the other hand, integrating variants that perturb enzyme functionality from the same RNA-seq data may enhance GEM accuracy, offering insights into genome-wide metabolic pathology. Our study pioneers the extraction of both transcriptomic and genomic data from the same RNA-seq data to reconstruct personalized metabolic models. We map genes with significantly higher load of pathogenic variants in Alzheimer's disease (AD) onto a human GEM together with the gene expression data. Comparative analysis of the resulting personalized patient metabolic models with the control models shows enhanced accuracy in detecting AD-associated metabolic pathways compared to the case where only expression data is mapped on the GEM. Besides, several otherwise would-be missed pathways are annotated in AD by considering the effect of genomic variants.
通过将基因表达数据映射到基因组规模代谢模型（GEMs）来生成特定疾病状态下的代谢模型，是从代谢角度阐释疾病机制的常规方法。另一方面，整合来自同一RNA测序数据中影响酶功能的变异，可能会提高GEM的准确性，从而深入了解全基因组范围的代谢病理学。我们的研究率先从同一RNA测序数据中提取转录组和基因组数据，以重建个性化代谢模型。我们将阿尔茨海默病（AD）中具有显著更高致病变异负荷的基因与基因表达数据一起映射到人类GEM上。与仅将表达数据映射到GEM上的情况相比，将所得个性化患者代谢模型与对照模型进行比较分析，结果表明在检测与AD相关的代谢途径方面准确性有所提高。此外，通过考虑基因组变异的影响，在AD中注释了一些原本可能被遗漏的途径。
