For centuries, aging was considered inevitable and immutable. Geroscience provides the conceptual framework to shift this focus toward a new view that regards aging as an active biological process, and the biological age of an individual as a modifiable entity. Significant steps forward have been made toward the identification of biomarkers for and measures of biological age, yet knowledge gaps in geroscience are still numerous. Animal models of aging are the focus of this perspective, which discusses how experimental design can be optimized to inform and refine the development of translationally relevant measures and biomarkers of biological age. We provide recommendations to the field, including: the design of longitudinal studies in which subjects are deeply phenotyped via repeated multilevel behavioral/social/molecular assays; the need to consider sociobehavioral variables relevant for the species studied; and finally, the importance of assessing age of onset, severity of pathologies, and age-at-death. We highlight approaches to integrate biomarkers and measures of functional impairment using machine learning approaches designed to estimate biological age as well as to predict future health declines and mortality. We expect that advances in animal models of aging will be crucial for the future of translational geroscience but also for the next chapter of medicine.

中文翻译：

---

与老年科学相关的动物模型：生物标志物的当前趋势和未来前景，以及生物衰老的测量


几个世纪以来，衰老被认为是不可避免且不变的。Geroscience 提供了一个概念框架，将这一重点转移到一种新的观点上，即将衰老视为一个积极的生物过程，而个体的生物年龄是一个可改变的实体。在识别生物年龄的生物标志物和测量方面已经取得了重大进展，但老年科学方面的知识差距仍然存在。衰老的动物模型是该观点的重点，它讨论了如何优化实验设计，以告知和完善生物年龄的转化相关措施和生物标志物的开发。我们为该领域提供建议，包括：纵向研究的设计，其中受试者通过重复的多层次行为/社会/分子分析进行深度表型分析;需要考虑与所研究物种相关的社会行为变量;最后，评估发病年龄、病理严重程度和死亡年龄的重要性。我们重点介绍了使用机器学习方法整合生物标志物和功能障碍测量的方法，这些方法旨在估计生物年龄以及预测未来的健康状况下降和死亡率。我们预计衰老动物模型的进步对于转化老年科学的未来以及医学的下一章都至关重要。