metabnorm

metabnorm applies a mixed model-based normalization to metabolomics data to estimate metabolite correlation matrices and assess effects of NMR calibration standardization for improved detection of true metabolite interactions.

Key Features:

• Mixed Model Approach: Employs a mixed model-based normalization that simultaneously estimates a correlation matrix for metabolite data.
• Robustness and Performance: Validated on simulated and real datasets, demonstrating improved identification of true correlations among metabolites.
• Impact of Standardization: Evaluates effects of standardizing NMR data using calibration standards on correlation estimation, reporting slight influences in simulation and minimal differences in real datasets.

Scientific Applications:

• Metabolic Network Correlation Analysis: Supports analysis of metabolite–metabolite interactions within metabolic networks by providing normalized data and estimated correlation matrices.
• NMR Metabolomics Standardization Studies: Enables assessment of how calibration-standard-based NMR data standardization affects correlation estimates in metabolomics studies.

Methodology:

Utilizes a mixed model framework to normalize metabolomics data, simultaneously estimates correlations between metabolites as part of the normalization, and assesses the influence of NMR calibration-standard standardization on correlation estimates via simulation and empirical analysis.