@inproceedings{Jendoubi:2020, author = {Jendoubi, T and Ebbels, TMD}, title = {Integrative analysis of time course metabolic data and biomarker discovery}, url = {http://arxiv.org/abs/1801.07767v2}, year = {2020} }
TY - CPAPER AB - Metabonomics time-course experiments provide the opportunity to understandthe changes to an organism by observing the evolution of metabolic profiles inresponse to internal or external stimuli. Along with other omic longitudinalprofiling technologies, these techniques have great potential to complement theanalysis of complex relations between variations across diverse omic variablesand provide unique insights into the underlying biology of the system. However,many statistical methods currently used to analyse short time-series omic dataare i) prone to overfitting or ii) do not take into account the experimentaldesign or iii) do not make full use of the multivariate information intrinsicto the data or iv) unable to uncover multiple associations between differentomic data. The model we propose is an attempt to i) overcome overfitting byusing a weakly informative Bayesian model, ii) capture experimental designconditions through a mixed-effects model, iii) model interdependencies betweenvariables by augmenting the mixed-effects model with a conditionalauto-regressive (CAR) component and iv) identify potential associations betweenheterogeneous omic variables . AU - Jendoubi,T AU - Ebbels,TMD PY - 2020/// TI - Integrative analysis of time course metabolic data and biomarker discovery UR - http://arxiv.org/abs/1801.07767v2 ER -