This thesis evaluates the ability of ¹H nuclear magnetic resonance (NMR) spectroscopic and ultra performance liquid chromatography-mass spectrometry (UPLC-MS) strategies to extract latent biochemical information from complex biofluid matrices from animal models and humans. ¹H NMR spectroscopy was applied to urine samples to investigate time-dependent biochemical changes in acid-base balance studies of metabolic alkalosis and metabolic acidosis and in a study of caloric restriction in order to establish normal physiological variation in organic acids including tricarboxylic acid (TCA) cycle intermediates. A method for profiling organic acids in human urine was developed, optimised and validated using UPLC-MS and then applied to targeted and untargeted metabolic profiling. A combined approach using both techniques was then applied in order to assess the relative efficiencies and merits of ¹H NMR spectroscopy and UPLC-MS for diagnosis of unknown inborn errors of metabolism (IEM) diseases using human urine samples in a blind fashion. Both NMR and UPLC-MS analytical platforms detected and identified all of the IEM correctly and revealed complementary biochemical information from IEM and controls. This evaluation was further extended and scaled up to assess the capability of the technologies for extracting latent biological information from urine samples collected in a large human population study. Chemometric methods such as Principal Component Analysis (PCA) and Orthogonal Projection to Latent Structure Discriminant Analysis (O-PLS-DA) were used to interpret and compare the two spectroscopic data types. These statistical methodologies allowed the detection of population specific biomarkers and the establishment of key metabolic phenotype differences in selected populations. In particular, both analytical approaches were able to discriminate Icelandic participants from other populations. The methods developed as part of this project should prove useful in widespread metabonomic applications in pharmaceutical and disease related areas.