A statistical procedure to selectively detect metabolite signals in LC-MS data based on using variable isotope ratios

Area of Science:

• Metabolomics
• Analytical Chemistry
• Computational Biology

Background:

• Stable isotope-labeled compounds aid metabolite signal tracing in LC-MS data.
• Current methods for mining metabolite signals in complex LC-MS datasets have limitations in filtering efficiency and confidence.

Purpose of the Study:

• To propose an enhanced statistical procedure for increasing metabolite signal mining efficiency and confidence in LC-MS data.
• To introduce the signal mining algorithm with isotope tracing (SMAIT) for improved metabolite detection.

Main Methods:

• Development of an in-house computational program, SMAIT, to perform a statistical procedure.
• Utilizing the correlation between varying concentration ratios of native/stable isotope-labeled compounds and their instrumental response ratio.
• Application of the method using di-(2-ethylhexyl) phthalate (DEHP) as an example in LC-MS data.

Main Results:

• The statistical procedure effectively filtered 15 probable metabolite signals from 3617 peaks in the LC-MS data.
• The identified signals were structurally related to DEHP, demonstrating the method's specificity.
• SMAIT significantly improved the confidence and efficiency of metabolite signal detection.

Conclusions:

• The proposed statistical procedure and SMAIT offer a highly confident approach to screen metabolite signals.
• This method facilitates the confident detection of probable metabolites from compound-derived precursors in complex LC-MS datasets.
• SMAIT represents a valuable tool for advancing metabolomic research and data analysis.