Yale maps shared cancer risk to regions, not the whole genome
Hu, Muheyati, Xu, DeWan and Zhao published a PLOS Genetics observational genetics study on 18 September 2026. SUPERGNOVA local scans across a sixteen-site cancer panel plus a pan-cancer phenotype yielded twenty significant genome-wide pairs and eighty-two regions among sixty-six pairs. Five mutually correlated clusters showed inflammatory enrichment; cross-cancer scores marked five pairs on immune routes. Overlap across the whole genome stays modest. No therapy is claimed.
Science··Midday
Twenty pairs, 82 regions
Hu and colleagues at Yale published an observational genetics study on 18 September 2026. They applied SUPERGNOVA locally across a sixteen-site cancer panel plus a pan-cancer phenotype. They report twenty significant pairs at genome scale and eighty-two regions spanning sixty-six pairs in the local analysis. The journal page is the only newsroom account located after a Yale News search returned none.[1]
Immune-pathway PRS, not a smear
Five genomic stretches formed mutually correlated cancer clusters with inflammatory-function enrichment. Cross-cancer scores flagged five pairs. Those links sat in pleiotropy groups and immune routes, not a smear across the whole genome.[1]
Genome-wide sharing stays modest
Overlap across the whole genome remains modest in this observational genetics paper; the authors do not state a therapy. Yale News search for Zhao and polygenic cancer returned no results, and a ScienceDaily search page did not carry this PLOS Genetics paper, so the card stays on the journal. Hu and colleagues stay inside the published abstract counts. A second newsroom page was not found on Yale News or ScienceDaily, so this card does not invent one. The modest genome-wide overlap is the authors own limit, not a clinic protocol.[1]