Fig. 2: SoL abundances are decreased in IBD metabolomics datasets and independent IBD patient samples.

a Box plots showing the relative abundance of SoL candidates detected in dataset 1 (upper) from non-IBD (red boxes, n = 124 individuals) and IBD individuals (blue boxes, n = 348 individuals) and in dataset 2 (lower) from non-IBD (red boxes, n = 56 individuals) and IBD individuals (blue boxes, n = 164 individuals). For each feature, the corresponding SoL is noted in the bottom label. The prefixes of metabolomic feature names correspond to the detection method used: In dataset 1, HILp indicates the HILIC-positive method and HILn indicates the HILIC-negative method. In dataset 2, HILIC-pos indicates the HILIC-positive method and HILIC-neg indicates the HILIC-negative method. Details for the corresponding LCMS methods can be found in the original studies. Significance was determined using the one-sided Wilcoxon rank sum test with the hypothesis that the abundance of SoL was higher in the non-IBD than in IBD group. Exact p-values from left to right are: 1.2E-11, 3.9E-13, 2.3E-14, 1.1E-6, 1.1E-8 in dataset 1 (upper) and 3.3E-8, 6.2E-9, 1.2E-7, 4.5E-10, and 6.5E-11 in dataset 2 (lower). b Box plots showing the absolute abundance of SoLs B, C, and F measured by targeted metabolomics in an independent cohort of IBD patient stool samples. SoL B and F were found to be significantly decreased in IBD (blue boxes, n = 40) compared to non-IBD (red boxes, n = 20) samples. All box plots include center lines representing the median, box limits representing upper and lower quartiles, whiskers representing the 1.5x interquartile range, and points representing outliers. Source data are provided in the Source Data file. Significance was determined using two-sided Student’s t-test. The exact p-values for SoL B and SoL F were 0.0241 and 0.0482, respectively. For all p values: *0.01 <p < 0.05, **0.001 <p < 0.01, and ***p < 0.001.