Fig. 10From: MORPHIOUS: an unsupervised machine learning workflow to detect the activation of microglia and astrocytesMORPHIOUS identifies proximal astrocytes in a mouse model amyloidosis. Proximal astrocytic activation clusters (orange line) (A) were observed in association with amyloid-beta plaques (B). Compared with nonTg and distal astrocytes, proximal astrocytes exhibited greater levels of GFAP immunofluorescence (C) and percent area coverage (D). Pearson correlation analysis demonstrates that proximal astrocyte cluster sizes correlated with amyloid plaque load (E). Proximal astrocytes exhibited increased levels of amyloid-beta plaque coverage (F) and plaque size (G). Images (A–B) were taken at 20X magnification. Between-group differences were assessed via a one-way ANOVA with Tukey’s post-hoc analysis. Correlations were assessed via linear regression analysis and the Pearson correlation coefficient (R2) is reported. Significance: * P < 0.05; ** P < 0.01; *** P < 0.001; **** P < 0.0001. Data represent means ± SD; N = 4 per group. Scale bar: 100 µm (A, B). GFAP, glial fibrillary acidic protein; nonTg, non-transgenic littermates of TgCRND8 mice; Prox., proximal; Tg, TgCRND8 miceBack to article page