Context, not contradiction: a pan-cancer framework for conforming and nonconforming co-expression programmes

PTK2 expression across 17 TCGA cancer cohorts is linked to poor outcome in some cancers and better outcome in others, independently of tumour purity.

• Why it matters: Conflicting prognostic reports may reflect lineage-specific co-expression rather than cohort noise, but pan-cancer analyses are biased by thresholding, dosage effects, and unmeasured cross-type conservation.
• What they did: The authors recomputed co-expression threshold-free per cohort, partitioned cis versus trans transcripts after resolving retired symbols, froze gene sets via a recurrence rule, used expression-decile-matched permutation nulls, and residualised copy number using the cis set as a positive control.
• The result: Cohorts separated into conforming versus nonconforming groups enriched for different biology; the 44-transcript trans programme persisted after copy-number adjustment while the cis control collapsed by 38%, and trans co-expression was recoverable in 133 of 243 single-cell cases.

AI summary of the abstract · may contain errors · not medical advice · read the paper.