Analysing delays between time course gene expression data and biomarkers
Jasmin Straube, Anne Bernard, Bevan Emma Huang, Le Cao K.-A. · 2015
Associating time course gene expression data to biomarkers can help to understand disease progression or response to therapy. However, detecting associations between these expression profiles is not a trivial task. Often expression changes occur not simultaneously but delayed in time and common used methods to detect correlation will fail to identify these associations. We have developed an efficient approach, DynOmics, based on Fast Fourier Transform to identify coordinated response dynamics between time course 'omics' experiments and specific biomarkers of interest while taking time shift into account. We applied DynOmics to a rat study investigating molecular response dynamics to different dosages of acetaminophen ('paracetamol'). We show how DynOmics can extract relevant molecule expression profiles that enables a better understanding of the molecular pathways related to acetaminophen toxic dosage and renal damage.