Ensemble polygenic risk scores built without individual-level training data, in Nature Genetics
anshulkundaje · x · 2026-11-18
A Nature Genetics paper introduces ensemble-based polygenic risk score construction that avoids using individual-level data for training, sidestepping privacy and data-sharing barriers in genetic risk prediction.
More from Research
- PANGEA atlas of 12.6M cells reveals shared inflammatory cell programs across organs and diseases — BoWang87 · 2026-11-18
- LLaVA-OneVision-2.5 released: first clear signal that codec input beats frame sampling — liuziwei7 · 2026-11-18
- Paper validates probabilistic safety judges for RL, finds harm hides in confidence blind spot — kalyan_kpl · 2026-11-18
- AlphaGenome 2 hits bioRxiv, with Gemini drafting the paper's methods and results sections — anshulkundaje · 2026-11-18
- From 1000-step DDPM to 1-step generation: a hands-on turbo diffusion walkthrough — RisingSayak · 2026-11-18
- Timelines collapse after METR: Christiano warns of AI takeover risk within 6-12 months — TheZvi · 2026-11-18