Extended Data Fig. 3: PSME4 is not strongly associated with other factors in Lung Adenocarcinoma.
a – d, Using the protein abundance of PSME4 normalized to the mean abundance of the core proteasome from the CPTAC LUAD cohort compared to patient data, we examined the association to other factors. (a) The significance of the association between each of the factors listed and the abundance of PSME4 across the samples. For discreet factors, the mean abundance between the different groups was compared with a one-way ANOVA. For continuous factors, the significance of the spearman correlation between the protein abundance and factor is displayed. (b) Slight increase in PSME4 abundance in tumors that do not have a KRAS mutation (two-sided student’s t test, n = 111 tumors). (c) Change in abundance of all the proteasome subunits between KRAS mutant and wild type KRAS tumors. As most proteasome subunits are decreased in abundance in KRAS mutant cancers, we conclude that the change we observe in PSME4 is not due specifically to PSME4 level or function. Significance determined from two-sided student’s T test, n = 111 tumors. (d) The abundance of PSME4 normalized to the core in each of the histological subtypes of PSME4 (n = 111 tumors). Variation between the subtypes only occurs in rare groups that cannot clearly be distinguished due to sample size. e-i, Quantification of PSME4 in NSCLC subtypes lung adenocarcinoma (LUAD) and squamous cell lung carcinoma (SQCLC) measured by immunohistochemistry staining. Samples are stratified by subtype (e), smoking level (f), T stage (g), N stage (h) or TNM stage (i). Dots are colored by subtype (LUAD – yellow, SQCLC - red). 50 independent patient samples were assessed. Box plots span the first to third quartiles. Significance determined by Wilcoxon rank sum test.
