重叠密度热图：代谢组学研究的新方法

Overlap Density Heatmaps: A Novel Approach to Metabolomics Research

Abstract

In this study, we introduce a novel tool, the Overlap Density Heatmap (ODH), for the visualization and quantitative evaluation
of  the  (dis)similarity  in  massive  amounts  of  spectral  or chromatographic data.  This new spectral visualization tool was applied to a study of the 1 H NMR spectra of human serum samples  from  37  diabetic  and  non-diabetic  subjects.  OD consensus spectra of the normal and diabetic samples were generated,  and  a  difference  spectrum  of  the  two  was determined. This difference spectrum identified diagnostic peak regions which can distinguish diabetic from normal patient samples. Using the difference spectrum as a search against the whole dataset revealed clear separation between both patient populations. Further, we then deployed the difference spectrum as a search query against a database of known metabolites. Only D-glucose, a known biomarker for diabetes, was retrieved. The result illustrates the versatility of applying the ODH to identifying the requisite metabolite/biomarker for a disease type such as diabetes.

In a complementary study, we deployed Principal Component Analysis (PCA) to the analysis of the same 37 samples. Good
class  separation  between  both  patient  populations  was  obtained. The PCA loadings plot, which highlights important peak  positions,  and  may  also  implicate  metabolites,  was compared to the difference spectrum from ODH and found to share many similar features. When used as a search query, the loadings plot spectrum retrieved a single hit, (D-glucose), from the metabolite database, similarly as was found for the ODH-based search query. These findings demonstrate that ODH can provide an unbiased approach to enhancing the multivariate analysis and interpretation of NMR-based metabolomics data.

Conclusions

We have shown that this new Overlay Heat Density (ODH) technique  has  wide  applications  in  spectral  visualization, analysis,  and  metabolomics  research.  In  combination  with chemometrics tools, the ODH technology offers a very powerful approach  to  the  study  of  metabolite  identification  and characterization. Incorporation of Infometrix' chemometrics software   (Pirouette ® )   into   the   KnowItAll   package   offers researchers  a  fully-integrated  NMR-based  metabolomics platform.  Such an integrated  platform opens the door to multi-technique metabolomics studies, which should provide more options for future research in this area.