Mathematical Framework for Quantifying Delocalization in MALDI-MSI via a Composite Scoring Approach

Matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI) has become a widely used tool for demonstrating the spatial distribution of biomolecules in tissue. One common issue in this technique that can have an impact on sensitivity and spatial resolution is analyte delocalization, in which the analyte spreads across the tissue or beyond the tissue boundaries, often due to sample handling or matrix application. In this study, we tested several metrics to evaluate delocalization using MALDI-MSI data from mouse brain sections. These metrics included the distance between the centers of mass of the nonzero-intensity pixels of tissue and global, the maximum and mean distances of off-tissue signal from the border, and the total area of background signal. After comparison of these metrics to the data set, a linear combination of area with mean distance was defined as the delocalization score. This score can be tuned depending on the application, and we found that a higher amount of weight on the area worked well in practice. Studying the Pearson correlation coefficients of the delocalization scores across the four groups of mice revealed a strong correlation among the analytes, suggesting that their delocalization patterns behave similarly across all groups.