Nonalcoholic fatty liver disease (NAFLD) is the most common chronic liver disease affecting people’s health worldwide. Exploring the potential biomarkers and dynamic networks during NAFLD progression is urgently important. Differentially expressed genes (DEGs) in obesity, NAFL and NASH were screened from GSE126848 and GSE130970, respectively. Gene set enrichment analysis of DEGs was conducted to reveal the Gene Ontology (GO) biological process in each period. Dynamic molecular networks were constructed by DyNet to illustrate the common and distinct progression of health- or obesity-derived NAFLD. The dynamic co-expression modular analysis was carried out by CEMiTool to elucidate the key modulators, networks, and enriched pathways during NAFLD. A total of 453 DEGs were filtered from obesity, NAFL and NASH periods. Function annotation showed that health-NAFLD sequence was mainly associated with dysfunction of metabolic syndrome pathways, while obesity-NAFLD sequence exhibited dysregulation of Cell cycle and Cellular senescence pathways. Nine nodes including COL3A1, CXCL9, CYCS, CXCL10, THY1, COL1A2, SAA1, CDKN1A, and JUN in the dynamic networks were commonly identified in health- and obesity-derived NAFLD. Moreover, CYCS, whose role is unknown in NAFLD, possessed the highest correlation with NAFLD activity score, lobular inflammation grade, and the cytological ballooning grade. Dynamic co-expression modular analysis showed that module 4 was activated in NAFL and NASH, while module 3 was inhibited at NAFLD stages. Module 3 was negatively correlated with CXCL10, and module 4 was positively correlated with COL1A2 and THY1. Dynamic network analysis and dynamic gene co-expression modular analysis identified a nine-gene signature as the potential key regulator in NAFLD progression, which provided comprehensive regulatory mechanisms underlying NAFLD progression.

中文翻译：

---

非酒精性脂肪肝的动态共表达模块网络分析

非酒精性脂肪性肝病 (NAFLD) 是影响全世界人们健康的最常见的慢性肝病。在 NAFLD 进展过程中探索潜在的生物标志物和动态网络非常重要。分别从 GSE126848 和 GSE130970 中筛选出肥胖、NAFL 和 NASH 中的差异表达基因 (DEG)。对DEGs进行基因集富集分析，揭示每个时期的基因本体（GO）生物学过程。动态分子网络由 DyNet 构建，以说明健康或肥胖衍生的 NAFLD 的常见和不同进展。CEMiTool 进行了动态共表达模块分析，以阐明 NAFLD 期间的关键调节剂、网络和富集途径。从肥胖、NAFL 和 NASH 时期共过滤了 453 个 DEG。功能注释显示健康-NAFLD序列主要与代谢综合征通路功能障碍有关，而肥胖-NAFLD序列则表现出细胞周期和细胞衰老通路的失调。动态网络中的 9 个节点，包括 COL3A1、CXCL9、CYCS、CXCL10、THY1、COL1A2、SAA1、CDKN1A 和 JUN，通常在健康和肥胖衍生的 NAFLD 中被识别。此外，CYCS 在 NAFLD 中的作用未知，与 NAFLD 活动评分、小叶炎症分级和细胞学气球样分级的相关性最高。动态共表达模块分析表明，模块 4 在 NAFL 和 NASH 中被激活，而模块 3 在 NAFLD 阶段被抑制。模块3与CXCL10呈负相关，模块4与COL1A2和THY1呈正相关。