Data correction, normalization and validation for quantitative high-throughput metabolomic profiling

摘要

Metabolomic profiling of a biological sample using a separation-molecular ID process, such as gas chromatography-mass spectrometry (“GC-MS”), requires the derivatization of the original sample. Quantitative GC-MS metabolomics is possible if the derivative is in one-to-one proportional relationship with the original concentration profile, wherein the proportionality remaining constant among samples. Two types of biases may be introduced into determination of a metabolomic profile to alter these conditions. The first type of bias is produced by a change in the proportionality size between profiles and is corrected by way of an internal standard. The second type of bias may distort the one-to-one relationship and change the proportionality between the profiles to a different fold-extent for each metabolite in a sample. The metabolomic profile data is corrected from these biases to reduce the risk of assigning biological significance to changes due only to chemical kinetics. A data correction and validation strategy provides for a weighted average of metabolite derivatives after derivatization of an original metabolite and before steady state equilibrium is established between plural metabolite derivatives to maintain high-throughput data acquisition and metabolomics analysis.