Fig. 1: Lipid abundance in control brains without brain pathology.
A Box and whisker plots displaying the z-scored (represents the distance from the population mean in units of standard deviations) sum of lipids per class and the distribution in the different regions (CA n = 13, CBM n = 14, CiC n = 13, FrC n = 12, PHp n = 11, PtC n = 15, PU n = 13, TCtx n = 16). The whiskers show the minimum and maximum and the boxes show the 25th percentile, the median and the 75th percentile. Values outside 1.5 times the interquartile are represented by dots. It was demonstrated that the phospholipids, ceramide and sphingomyelin displayed an even distribution between the different regions. B Hierarchical clustering showing the average of each compound per region in controls. Three main clusters formed consisting of (i) frontal cortex, caudate and parahippocampus, (ii) cerebellum, cingulate cortex and putamen, and (iii) parietal and temporal cortices. The primary driving parameter for this clustering was found to be an elevated abundance of most lipids in the frontal cortex-caudate-parahippocampus cluster. The clustering method was set to average, with cosine as distance metric. C UMAP projection with HDBSCAN clustering, demonstrating how the lipid classes relate to each other. The projection grouped into six main clusters, largely made up of lipids from the same class although overlap between classes was also observed. The UMAP points were coloured according to main class and shaped according to sub-class. The projection was set to model 12 neighbours (average lipid class size) with correlation as the distance metric. Source data are provided as a Source Data file.
