Q-omics provides the consensus-scored KRAS profile across patient tissues and cancer cell-line models. KRAS expression is associated with patient survival in 24 of 34 cancer types, with the highest sampling consensus in CESC. Among the 18 cancer types available for tumor–normal comparison, KRAS is differentially expressed in 13, with the highest sampling consensus in COAD. Additionally, KRAS protein abundance shows 26,841 significant protein co-abundance associations, with the highest sampling consensus in GBM. Together, these results highlight CESC, COAD, and GBM as cancer lineages where KRAS shows reproducible signals across survival, tumor–normal expression, and patient cross-omics analyses.
Every result is evaluated using two consensus scores. Sampling consensus measures how consistently a finding is reproduced within a cancer lineage across different conditions. Lineage consensus measures how broadly the result is shared across cancer types, distinguishing pan-cancer signals from lineage-specific patterns.
Premium analyses for KRAS — synthetic lethality, tumor antigen, and pembrolizumab response.
This table summarizes KRAS survival associations across molecular data types. KRAS RNA expression shows survival associations in the most cancer types (24), followed by mutation status (14) and mass-spec protein abundance (9). The rightmost column indicates the cancer type with the highest sampling consensus for each molecular layer.
This table ranks reproducible KRAS RNA expression–survival associations across cancer types. High KRAS expression shows unfavorable associations in CESC, ACC, MESO, PAAD and UVM, but favorable associations in KIRC. The CESC Kaplan–Meier curve shows clear separation, with the high-expression group declining faster, consistent with the unfavorable association (log-rank p < 0.001). Together, the overview and detailed table identify CESC as the clearest survival context for KRAS RNA expression.
This table summarizes KRAS tumor–normal expression differences by data type. RNA shows broader differences across cancer types, with a lineage consensus of 13, while mass-spec protein shows differences in 12. The strongest signals are observed in COAD for RNA and CCRCC for protein.
This table ranks reproducible tumor–normal expression differences for KRAS. A negative fold-change indicates higher expression in normal tissue than in tumor tissue. KRAS shows lower tumor expression in COAD and THCA and higher tumor expression in BRCA, CHOL, LUAD and LUSC. The COAD box plot shows higher KRAS RNA expression in normal versus tumor tissue (log2 FC = −0.581, t-test p < 0.001).
This table shows molecular features associated with KRAS in patient tissues and cancer cell lines. In patient samples, KRAS shows the broadest associations at the RNA and protein expression levels, with GBM recurring as the lineage with the largest associated feature set. In cancer cell lines, KRAS RNA and mutation anchors are most strongly linked to RNA-expression features, especially in PANCREAS, while CRISPR and shRNA rows add functional-dependency signals in BLOOD_Leukemia and UPPER_AERODIGESTIVE_TRACT.
