TY - JOUR
T1 - Clustering gene expression time series data using an infinite Gaussian process mixture model
AU - McDowell, Ian C.
AU - Manandhar, Dinesh
AU - Vockley, Christopher M.
AU - Schmid, Amy K.
AU - Reddy, Timothy E.
AU - Engelhardt Martin, Barbara
N1 - Funding Information:
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 grant 5T32GM071340. AKS was funded by National Science Foundation MCB 1417750 and NSF CAREER 1651117. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. We thank Alejandro Barrera who provided insight into packaging the DPGP software. We thank our colleagues at Duke and Princeton Universities for insightful conversations about this research.
Publisher Copyright:
© 2018 McDowell et al.
PY - 2018/1
Y1 - 2018/1
N2 - 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.
AB - 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.
UR - http://www.scopus.com/inward/record.url?scp=85041379913&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85041379913&partnerID=8YFLogxK
U2 - 10.1371/journal.pcbi.1005896
DO - 10.1371/journal.pcbi.1005896
M3 - Article
C2 - 29337990
AN - SCOPUS:85041379913
SN - 1553-734X
VL - 14
JO - PLoS computational biology
JF - PLoS computational biology
IS - 1
M1 - e1005896
ER -