Extended Data Fig. 4: pLGG show immunosuppressive myeloid populations, as related to Fig. 3.

a. Proportion of BMDM and MG subtypes positive for the activation marks CD206 (n = 120). Samples devoid of these populations were excluded from this analysis. Data are presented as mean, all points plotted. b. Expression of HAVCR2, LGALS9, and PCD1 using the Open Pediatric Brain Tumor Atlas (Boxplots depict the median, first and third quartiles. Whiskers extend from the hinge to the largest/smallest value no further than 1.5 * interquartile range from the hinge). c. Percentage of HAVCR2 (TIM-3) positive myeloid and lymphoid cells, in pLGG and H3K27M-mutant tumors (Andrade et al). d. Percentage of HAVCR2 (TIM-3) positive cells in pLGG immune populations (Andrade et al). e. A representative immunofluorescence image of co-stain using antibodies for PD1 or TIM3 and CD68. Merge depicted in a human tonsil sample used as positive control for both markers. Magnification: 20x. f. Kaplan-Meier curve depicting the association between TIM-3 (HAVCR2) gene expression and PFS in the whole cohort with survival and TIM3 data (n = 63) or in patients at risk of progression (that is non-GTR patients only – n = 51); p-values based on Log-Rank (non-corrected for multiple testing). g. Circle plots for the corresponding interacting pair modules where the HAVCR2-Gal9 pair was enriched. Cell types are shown in the outer circle, while interacting pairs are displayed in the inner one. Directional links originate from ligands and are pointing towards receptors; they are colored by a scaled (−2,2) interaction score to highlight relative differences in expression. h. Expression of C3 and C3AR1 in immune populations by scRNAseq, in H3K27M-mutant tumors (Andrade et al). i. Differentially expressed genes between TIM-3 positive and negative microglia. Genes with an adjusted p value < 0.05 and log2 fold change > 0.5 are shown in red/blue (two-sided Wilcoxon ranked sum test).

Source data