Shape-constrained, changepoint additive models for time series omics data with cpam.

Time series omics experiments are critical for studying a range of biological processes, such as cell differentiation and developmental programs or responses to pathogens and environmental cues. While statistical tools for differential analysis across static conditions have matured, a comparable comprehensive methodology is lacking for time series data. Here, we introduce cpam, a novel method and user-friendly R package that performs temporal differential analysis of omics time series data, including pseudobulked single-cell sequence data. Powerful features include change-point detection and shape-constrained temporal trend estimation to allocate omics data into similar clusters. The software handles case-only and case-control designs, incorporates quantification uncertainty, and provides an interactive interface with customisable visualisations, offering graphical and statistical insight into molecular processes. Performance evaluation shows that cpam outperforms existing time series methods in terms of control of the false discovery rate versus power to detect temporal changes and accurate changepoint estimation. Application to published data illustrates RNA isoform-level modelling with high-resolution clustering during human embryogenesis and the identification of 910 novel genes that respond to excess-light in Arabidopsis.