Frequent shift work causes disruption of the circadian rhythm and might on the long-term result in increased health risk. Current biomarkers evaluating the presence of circadian rhythm disturbance (CRD), including melatonin, cortisol and body temperature, require 24-hr ("around the clock") measurements, which is tedious. Therefore, these markers are not eligible to be used in large-scale (human) studies. The aim of the present study was to identify universal biomarkers for CRD independent of time of day using a transcriptomics approach. Female FVB mice were exposed to six shifts in a clockwise (CW) and counterclockwise (CCW) CRD protocol and sacrificed at baseline and after 1 shift, 6 shifts, 5 days recovery and 14 days recovery, respectively. At six time-points during the day, livers were collected for mRNA microarray analysis. Using a classification approach, we identified a set of biomarkers able to classify samples into either CRD or non-disrupted based on the hepatic gene expression. Furthermore, we identified differentially expressed genes 14 days after the last shift compared to baseline for both CRD protocols. Non-circadian genes differentially expressed upon both CW and CCW protocol were considered useful, universal markers for CRD. One candidate marker i.e. CD36 was evaluated in serum samples of the CRD animals versus controls. These biomarkers might be useful to measure CRD and can be used later on for monitoring the effectiveness of intervention strategies aiming to prevent or minimize chronic adverse health effects.