ZeitZeiger is a method for regularized supervised learning on high-dimensional data from an oscillatory system. ZeitZeiger relies on periodic smoothing splines and sparse principal components. ZeitZeiger can quantify rhythmic behavior, make accurate predictions, identify major patterns and important features, and detect when the oscillator is perturbed.
LimoRhyde: a flexible approach for differential analysis of rhythmic transcriptome data, Singer and Hughey, J Biol Rhythms 2018
Population-level rhythms in human skin with implications for circadian medicine, Wu et al., PNAS 2018
Evidence for widespread dysregulation of circadian clock progression in human cancer, Shilts et al., PeerJ 2018
Machine learning identifies a compact gene set for monitoring the circadian clock in human blood, Hughey, Genome Med 2017
Differential phasing between circadian clocks in the brain and peripheral organs in humans, Hughey and Butte, J Biol Rhythms 2016
ZeitZeiger: supervised learning for high-dimensional data from an oscillatory system, Hughey et al., Nucleic Acids Res 2016