Figure EV2. Quality control of cell type‐resolved mouse brain secretome resource.

1. Data transformation of the hiSPECS DIA secretome analysis of the brain cell types. Log₂ transformation of the intensities (N = 1). The central band indicates the median, the boxes indicate the lower or upper quartile, the whiskers indicate minimum or maximum, and dots indicate outliers.
2. Functional annotation clustering with DAVID 6.8 (da Huang et al, 2009a; da Huang et al, 2009b) for gene ontology term cellular component (FAT) of the 995 hiSPECS proteins identified using the mouse proteome as background. The dot sizes indicate the log₂ enrichment score.
3. Fold change in cell type‐specific proteins in the brain cells. Log₂ ratio of the average abundance in the specific cell type to the median abundance in the other cell types in the hiSPECS secretome or lysate analysis (Sharma et al, 2015) is shown. Known cell type‐specific marker proteins are highlighted for each cell type which reveals a strong enrichment in the lysate and secretome of the primary brain cells verifying the quality and comparability of the primary cultures. For example, the ectodomain of the membrane protein NCAM2 (with an y‐axis value of 3 in the log₂ scale) is secreted about eightfold more from oligodendrocytes compared with the median of the other three cell types and also expressed at a higher level in this cell type compared with the median of the other three cell types. Horizontal lines indicate the mean.
4. Principal component analysis (PCA). The secretomes of the four cell types segregated based on the two major components of all 995 proteins identified in at least 5 biological replicates in one cell type, which accounted for 44.9% and 19% of the variability, respectively.