Aging increases cell-to-cell transcriptional variability upon immune stimulation

Martinez-Jimenez et al. 2017.

Aging increases cell-to-cell transcriptional variability upon immune stimulation. Science 355, 1433–1436.

With proper single-cell RNA-seq. Martinez-Jimenez (from Marioni and Odom's groups, EMBL-EBI and CRUK Cambridge) profiled naive and activated CD4+ T cells from young and old mice.

Key findings:
(1) at baseline, young and old T cells looked similar;
(2) under LPS or anti-CD3 stimulation, old cells showed markedly increased cell-to-cell transcriptional heterogeneity, especially in immune-response genes;
(3) the increased variability was not random across the genome but concentrated in genes with cell-type-specific regulatory architecture.

This shows that computational drift is revealed by perturbation and localised to regulated loci, which is exactly what cybernetic attractor decay predicts (attractor erosion shows up under stress, not at rest, and concentrates at boundary regions).

Worth noting. This paper interprets the variance increase as stochastic noise.

Covolution framework reinterprets it as attractor flattening: the noise rises because the basin walls have eroded, not because the noise source has changed.