Q-omics provides the consensus-scored GPI profile across patient tissues and cancer cell-line models. GPI expression is associated with patient survival in 25 of 34 cancer types, with the highest sampling consensus in LUAD. Among the 18 cancer types available for tumor–normal comparison, GPI is differentially expressed in 17, with the highest sampling consensus in LUAD. Additionally, GPI protein abundance shows 22,078 significant protein co-abundance associations, with the highest sampling consensus in PDAC. Together, these results highlight LUAD, and PDAC as cancer lineages where GPI 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 GPI — synthetic lethality, tumor antigen, and pembrolizumab response.
This table summarizes GPI survival associations across molecular data types. GPI RNA expression shows survival associations in the most cancer types (25), followed by mutation status (9) and mass-spec protein abundance (8). The rightmost column indicates the cancer type with the highest sampling consensus for each molecular layer.
This table ranks reproducible GPI RNA expression–survival associations across cancer types. High GPI expression shows unfavorable associations in LUAD, UVM, HNSC, LGG, THCA and SKCM. The LUAD 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 LUAD as the clearest survival context for GPI RNA expression.
This table summarizes GPI tumor–normal expression differences by data type. RNA shows broader differences across cancer types, with a lineage consensus of 17, while mass-spec protein shows differences in 8. The strongest signals are observed in KIRC for RNA and CCRCC for protein.
This table ranks reproducible tumor–normal expression differences for GPI. A negative fold-change indicates higher expression in normal tissue than in tumor tissue. GPI shows higher tumor expression in LUAD, KIRP, KIRC, COAD, HNSC and LUSC. The LUAD box plot shows higher GPI RNA expression in tumor versus normal tissue (log2 FC = +1.451, t-test p < 0.001).
This table shows molecular features associated with GPI in patient tissues and cancer cell lines. In patient samples, GPI shows the broadest associations at the RNA and protein expression levels, with PDAC recurring as the lineage with the largest associated feature set. In cancer cell lines, GPI RNA and mutation anchors are most strongly linked to RNA-expression features, especially in BLOOD_Leukemia, while CRISPR and shRNA rows add functional-dependency signals in OVARY and UPPER_AERODIGESTIVE_TRACT.
