Clustering gene expression time series data using an infinite Gaussian process mixture model

Abstract

Transcriptome-wide time series expression profiling is used to characterize the cellular response to environmental perturbations. The first step to analyzing transcriptional response data is often to cluster genes with similar responses. Here, we present a nonparametric model-based method, Dirichlet process Gaussian process mixture model (DPGP), which jointly models data clusters with a Dirichlet process and temporal dependencies with Gaussian processes. We demonstrate the accuracy of DPGP in comparison to state-of-the-art approaches using hundreds of simulated data sets. To further test our method, we apply DPGP to published microarray data from a microbial model organism exposed to stress and to novel RNA-seq data from a human cell line exposed to the glucocorticoid dexamethasone. We validate our clusters by examining local transcription factor binding and histone modifications. Our results demonstrate that jointly modeling cluster number and temporal dependencies can reveal shared regulatory mechanisms. DPGP software is freely available online at https://github.com/PrincetonUniversity/DP_GP_cluster.

Author summary

Transcriptome-wide measurement of gene expression dynamics can reveal regulatory mechanisms that control how cells respond to changes in the environment. Such measurements may identify hundreds to thousands of responsive genes. Clustering genes with similar dynamics reveals a smaller set of response types that can then be explored and analyzed for distinct functions. Two challenges in clustering time series gene expression data are selecting the number of clusters and modeling dependencies in gene expression levels between time points. We present a methodology, DPGP, in which a Dirichlet process clusters the trajectories of gene expression levels across time, where the trajectories are modeled using a Gaussian process. We demonstrate the performance of DPGP compared to state-of-the-art time series clustering methods across a variety of simulated data. We apply DPGP to published microbial expression data and find that it recapitulates known expression regulation with minimal user input. We then use DPGP to identify novel human gene expression responses to the widely-prescribed synthetic glucocorticoid hormone dexamethasone. We find distinct clusters of responsive transcripts that are validated by considering between-cluster differences in transcription factor binding and histone modifications. These results demonstrate that DPGP can be used for exploratory data analysis of gene expression time series to reveal novel insights into biomedically important gene regulatory processes.