🌡️ Tumor expression vs normal tissue (GTEx)

All three RNA samples — the two tumors (hand primary, lung metastasis) and a normal-lung baseline — were reprocessed through the GTEx V11 pipeline (STAR 2.7.11b → RNA-SeQC 2.4.3, collapsed GENCODE v47), so their gene TPMs are directly comparable to GTEx normal-tissue expression. Each scatter plots one sample (x) against GTEx (y) — one point per gene, on a log₂(TPM + 1) scale. Genes far below the diagonal are higher in that sample than in any normal tissue; the driver genes EWSR1 and KLF15 are highlighted in red. The normal lung is an in-patient baseline: it should track GTEx lung, and a gene that is high in a tumor but not in the normal is tumor-specific rather than lung lineage.

How to use: on the Tumor vs GTEx tab, toggle the GTEx y-axis between the max or mean across 68 tissues, or percentile — what fraction of the ~19,800 GTEx normal samples each sample exceeds for that gene (genes near 100% are over-expressed relative to essentially all normal tissue). Search a gene or click any point to open an information panel for it and see its distribution across every GTEx normal tissue and every PCAWG tumor type below, with vertical lines marking where the three samples fall. Use the compartment chips to restrict the plots and the gene table (below the histograms) to a subcellular location — plasma-membrane genes are the most tractable immunotherapy targets. The gene table also reports a z-score — how many standard deviations each sample's log₂ expression sits above the GTEx (normal) or PCAWG (tumor) mean — which keeps discriminating genes even where the percentile saturates at 100%, and the full table downloads as TSV. The Assay Comparison tab plots all three sample pairs side by side — hand vs lung, and each tumor vs the normal lung — so you can see how correlated the tumors are and separate tumor-specific over-expression from shared lung-lineage expression.

Contact Us

Please contact us with any questions or comments.