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-theart 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 summaryTranscriptome-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 Data Availability Statement:All raw RNA-seq files are available from the Gene Expression Omnibus database (accession number GSE104714).Funding: BEE was funded by National Institutes of Health R00 HG006265, National Institutes of Health R01 MH101822, National Institutes of Health U01 HG007900, and a Sloan Faculty Fellowship. CMV, ICM, and TER were funded by National Institutes of Health U01 HG007900. CMV was also funded by National Institutes of Health F31 HL129743. DM was funded by National Institutes of Health training 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...

Brief Report: Cell Cycle Dynamics of Human Pluripotent Stem Cells Primed for Differentiation

“…E). We next used the nonparametric Dirichlet process Gaussian process mixture model to cluster fold changes in aligned reads (normalized to transcripts per million [TPM]) to assess changes in gene expression patterns associated with cell cycle progression. A total of 2,972 differentially expressed genes (false discovery rate (FDR) < 0.05) underwent clustering and 10 clusters emerged (Fig.…”

Cloning and functional analysis of JnCYCD3;1 in Jatropha nigroviensrugosus

“…A correlation circle map was created using the R circlize package [32]. The clustering of expression trends was performed using Python DP_GP [33].…”

Cell cycle dynamics of human pluripotent stem cells primed for differentiation

“…Using principal component analysis (PCA), we found that the strongest source of variation was in treatment vs control (PC1), followed by phase of the cell cycle (PC2) ( Figure 1E). We next used the nonparametric Dirichlet process Gaussian process mixture model (DPGP) 26 to cluster fold changes in aligned reads (normalized to transcripts per million (TPM)) to assess changes in gene expression patterns associated with cell cycle progression. A total of 2972 differentially expressed genes (FDR < 0.05) underwent clustering and 10 clusters emerged ( Figure 1F) with genes upregulated or downregulated in response to phase of the cell cycle following the DMSO treatment.…”

“…Differential genes underwent clustering with the Dirichlet process Gaussian process mixture model DPGP 26 software. 1000 iterations of clustering were performed with default software parameters.…”