Skin aging is a complex, multifactorial process driven by intrinsic biological mechanisms and environmental exposures, resulting in progressive functional and structural decline. Chronological age does not adequately capture this variability, highlighting the need for molecular biomarkers that reflect biological aging. In this context, transcriptomic aging clocks have emerged as a promising approach, as gene-expression profiles provide a dynamic representation of cellular and tissue states. This narrative review is based on a targeted literature search in PubMed and IEEE Xplore and focuses on transcriptomic aging clocks in human skin, with emphasis on gene-expression signatures, key biological pathways, and computational modeling strategies. These models consistently capture coordinated alterations in processes such as cellular senescence, DNA damage response, inflammation, and extracellular matrix remodeling. Representative transcriptomic frameworks, including models such as SkinAGE, illustrate the ability of gene-expression-based approaches to quantify biologically meaningful and dynamic aging states in the skin. Advances in machine-learning approaches, including deep learning and pathway-guided models, are critically evaluated, alongside the role of explainable artificial intelligence in enhancing model transparency and biological interpretability. Future developments are expected to integrate multi-omics data and digital twin frameworks, enabling the transition from static biomarkers toward dynamic, predictive, and personalized models of skin aging
Kefala et al. (Fri,) studied this question.