Gene expression measurements taken over multiple time points are valuable for describing dynamic biological phenomena such as an organ's response to injury or a tumor responding to therapy. However, such phenomena typically involve multiple biological processes occurring in parallel, making it difficult to identify and discern their respective contributions at any given time point.

In our lab, we use unsupervised machine learning to deconvolve time-series gene expression data into their underlying temporal components, with potential applications in the early detection and monitoring of diseases and recovery processes.