Age-related macular degeneration (AMD), a leading cause of irreversible vision loss in the elderly, is characterized by progressive degeneration of the retinal pigment epithelium and photoreceptors (Thomas et al., 2021). This ocular disorder predominantly impacts adults over 55 years of age, damaging the macular region and impairing essential visual functions such as reading, driving, and facial recognition, significantly reducing quality of life (Guymer and Campbell, 2023; Mitchell et al., 2018). AMD has emerged as a global health challenge due to aging populations and increased life expectancy. As a multifactorial disease, its pathogenesis involves genetic, environmental, and metabolic factors (Nowak, 2006), with recent evidence highlighting the role of gut microbiota dysbiosis (Zysset-Burri et al., 2023).

The gut-retina axis suggests that gut microbiota influences retinal health through immune modulation, metabolic regulation, and systemic inflammation (Chaiwiang and Poyomtip, 2019; Yang et al., 2024; Scuderi et al., 2021). Specific gut bacteria have been linked to chronic inflammatory states that may exacerbate retinal degeneration (Tîrziu et al., 2024; Larsen et al., 2023). Emerging research reveals distinct microbial compositional variations associated with ocular health, where certain bacterial taxa exhibit therapeutic potential, while others promote pro-inflammatory pathways linked to AMD (Li and Lu, 2023; Mao et al., 2023). However, these studies often lack comprehensive bioinformatics analysis to pinpoint specific genes and pathways involved. Furthermore, the role of differentially expressed gut microbiota-related genes (GMRGs) in AMD remains underexplored, limiting our understanding of the molecular mechanisms.

For the first time, this study systematically connects gutMGene-derived genes with AMD pathogenesis using a multi-algorithm machine learning framework. We aim to bridge this gap by analyzing differential GMRG expression in AMD patients and investigating their functional pathways. We utilized the gutMGene database to identify 238 GMRGs and performed differential expression analysis using the GSE29801 dataset. Enrichment analysis was conducted to highlight the biological significance of differentially expressed GMRGs. By applying machine learning algorithms and developing a predictive nomogram, we provide novel biomarkers for AMD and present new insights into the gut-retina axis. Fig. 1 presents the flowchart of the study. Our findings provide novel insights into the gut-retina axis in AMD, identifying key GMRGs and their pathways, which may guide future diagnostic and therapeutic strategies. The nomogram offers clinical utility for AMD risk prediction and patient management.