Metabolomics is an approach in which the profiles of metabolites in different tissues and/or biofluids are investigated to understand the changes induced following a modulation. We used this approach to investigate the biochemical effects of α-tocopherol in the liver using a rat model. Rats (21-day-old) were fed either an α-tocopherol-sufficient control (n = 10) or an α-tocopherol-deficient (n = 10) diet for 2 months before sacrifice. Livers were homogenized in methanol–chloroform–water (3 : 1 : 1, v/v/v), and the polar phase extracts of the liver samples were analyzed using 1H NMR. Multivariate statistical analysis of the data was performed using principal component analysis and orthogonal partial least squares-discriminant analysis. Identification of 1H NMR signals was performed primarily using the Human Metabolome Database, Biological Magnetic Resonance Data Bank and previous literature, and confirmed by spiking with metabolites and applying two-dimensional NMR. The statistical analysis revealed that α-tocopherol deficiency caused an increase in carnitine, choline, L-valine, L-lysine, tyrosine and inosine content and a reduction in glucose and uridine 5′-monophosphate content. Changes in carnitine and glucose suggest a possible shift in energy metabolism. Copyright © 2010 John Wiley & Sons, Ltd.