Fig. 3: GLASS-AI analysis of mouse models of LUAD.

a–e A cohort of Kras;^G12D/+ Rosa^mG/mG (K) (n = 11) and TAp73;^∆td/∆td Kras^G12D/+ (TK) (n = 13) mice were analyzed by GLASS-AI (a). After tumor segmentation and grading, the number of lung tumors (b) and tumor burden (c) for each mouse were totaled and analyzed by two-tailed Student’s t-test with Holm-Šídák correction for multiple comparisons. Each tumor was assigned a single grade based on the highest grade present that comprised ≥10% of the tumor’s area. Boxplots are presented in Tukey style; a line at the median with IQR, crosses indicating the mean, whiskers showing the lesser of 1.5× IQR or most extreme values, and points indicating outliers > 1.5× IQR from the median (b, c). The distribution of individual tumor sizes was examined by calculating the cumulative frequency of all grades of tumors in each genotype and was analyzed using a two-tailed Komolgorov-Smirnov test (d). Individual tumors were grouped by genotype and grade to perform a more in-depth size distribution analysis and analyzed by one-way ANOVA with two-tailed Tukey’s post-hoc test after log transforming the tumor area. Lines indicate median values, and whiskers show IQR (e). *p < 0.05 for the indicated comparison between K and TK mice (b, c, e). Numerals indicate *p < 0.05 between the noted grade within the same genotype (e).