Volume 2 Issue 8, August 2022

Mouse frailty can be measured with a frailty index by manually counting health deficits. Vivek Kumar and colleagues use machine learning to extract physical performance deficits from video data to create a ‘visual frailty index’. This automated technique may facilitate high-throughput research into new frailty interventions.

Using single-cell whole-genome sequencing, we identified and characterized the landscape of somatic single-nucleotide variants (sSNVs) in single cardiomyocytes from individuals across the human lifespan. Aged cardiomyocytes were found to have a higher burden of sSNVs and show mutational signatures that suggest failed repair of oxidative DNA damage.

Deep learning was applied to cellular images to predict senescence on the basis of nuclear morphology. These methods recognize senescence in diverse cell types, show increasing senescence with age in liver and dermis, and suggest that higher rates of senescence associate with several age-related diseases but reduced cancer risk.

This Perspective describes the blueprint, challenges and potential solutions for the transformation of Alzheimer’s disease clinical care pathway with biomarker-guided and digitally facilitated detection and intervention at early disease stages.

Goldman et al. demonstrate that meningeal lymphatic vessels play a role in sickness behavior. The authors also find that age-related lymphatic dysfunction increases susceptibility to sickness and that enhancement of meningeal lymphatic function improves movement in sick, aged mice.

Somatic mutations accumulate with age; however, the role they have in cardiac aging is unclear. Here Choudhury et al. describe the somatic mutation landscape of human heart muscle cells by single-cell whole-genome sequencing and report mutational signatures indicative of increased oxidative DNA damage and failed repair.

A new population of dysfunctional astrocytes in the aging mouse hippocampus called autophagy-dysregulated astrocytes (APDAs) show impaired protein homeostasis and defective regulation of synapse formation and elimination and appear early in a mouse model of Alzheimer’s disease.

Senescent cells are typically identified by a combination of senescence-associated markers, and the phenotype is heterogeneous. Here, using deep neural networks, Heckenbach et al. show that nuclear morphology can be used to predict cellular senescence in images of tissues and cell cultures.

The authors introduce a high-throughput machine-learning-based visual frailty index for mice that operates on video data from an open-field assay. The machine-vision-based approach extracts various morphometric and behavioral features from video to model frailty score and age.