Processing of mass spectrometry based metabolomics data for large scale screening studies and diagnostics
(English)Manuscript (preprint) (Other academic)
In mass spectrometry based metabolomics predictive data processing and sample classification based on representative sample subsets makes it possible to screen large sample banks or data sets in an efficient fashion regarding both data quality and processing time. This is a requirement for making use of high sensitivity and complexity metabolite data and to turn the metabolomics field into a competitive omics platform for biological interpretation and diagnostics. Predictive metabolomics by means of hierarchical multivariate curve resolution (H-MCR) followed by orthogonal partial least squares discriminant analysis (OPLS-DA) was used for the processing and classification of gas chromatography/time of flight mass spectrometry (GC/TOFMS) data characterizing human blood serum samples collected in a study of strenuous physical exercise. The efficiency of the predictive processing as a high throughput tool for generating high quality data is clearly proven and stated as a main benefit of the method. Extensive model validation schemes by means of cross validation and external predictions verified the robustness of the extracted systematic patterns in the data. Comparisons regarding the extracted metabolite patterns between models emphasized the reliability of the methodology in a biological information context. Furthermore, the high predictive power concerning longitudinal predictions provided proof for the diagnostic potential of the methodology. Finally, the predictive metabolite pattern was interpreted physiologically as well as verified in the literature, highlighting the biological relevance of the diagnostic pattern. The suggested approach makes it feasible to screen large data or sample sets with retained data quality and interpretation and to do this in a high throughput fashion. The method could be of value for sample bank mining, metabolome-wide association studies, verification of marker patterns and development of diagnostic systems.
Other Medical Sciences not elsewhere specified
IdentifiersURN: urn:nbn:se:umu:diva-50969OAI: oai:DiVA.org:umu-50969DiVA: diva2:471637