Fig. 2: Construction of the NELDI-MS platform for metabolic analysis.

a Illustration of the NELDI-MS platform. The lower left digital image showed the on-chip microarray of sample spots after the ferric matrix printing (Scale bar 5 mm). b High-angle annular dark field (HAADF) and elemental mapping analysis for the ferric nanoparticles (NPs) with Fe in red and O in green. The scale bar was 100 nm. c, d Detection for standard metabolites (concentrations of 1 mg mL⁻¹) of glutamate (Glu), lysine (Lys), arginine (Arg), glucose (Glc), and mannitol (Man) using the NELDI-MS or the organic matrix-assisted LDI-MS. The data was acquired in eight independent tests. The coefficients of variation (CVs) of the NELDI-MS detection for those standard metabolites were in (c). The MS intensity comparison of those standard metabolites between the NELDI-MS (using the ferric nanoparticles, NPs) and organic matrix-assisted LDI-MS (using 2,5-dihydroxybenzoic acid, DHB, or using α-cyano-4-hydroxy-cinnamic acid, CHCA) was in (d). The data in d was expressed as means ± standard errors and p-values in the analysis of variance (ANOVA) with two-sided Tukey post-hoc tests were shown. e The distribution and median of CVs of mass to charge ratio (m/z) features extracted from the representative mixture tear fluid and AH samples by the NELDI-MS. The line segments indicated quartiles and medians. The CVs were calculated by the data from eight independent tests and the representative mixture samples were prepared by mixing six ARC and six DC samples to include all possible m/z features. f Typical MS spectra of tear fluid samples and AH samples in the ARC and DC groups at m/z of 100–400. Source data are provided as a Source data file.