TimeSignature is a machine-learning approach to infer physiological time based on gene expression in human blood. A powerful feature is TimeSignature’s generalizability, enabling it to be applied to samples from disparate studies and yield highly accurate results despite systematic differences between the studies. It is able to pinpoint an individuals’ circadian phase to within 1.5hrs on average, and can be used with data from microarrays, RNA-seq, or PCR. It accuracy and universality is unique among expression-based predictors and addresses a major challenge in the development of reliable and clinically useful biomarker tests.
