Genome-wide association study

A control room wall covered in a million smoke-detector lights, no bulb singled out in advance. 1. Every light on the board gets checked, with no bulb picked out ahead of time -> GWAS scans common variants across the whole genome, hypothesis-free — no gene assumed in advance. 2. The chief only trusts a light if it clears a threshold so strict that barely any light could trip it by chance among a million -> GWAS uses a strict genome-wide significance threshold because so many independent variants are tested at once. 3. A tripped light only ever points to a block of buildings sharing one circuit, never to a single building -> Linkage disequilibrium: a significant signal marks a genomic region, not a specific causal gene. 4. It takes alarms from the whole city, not one street, before the control room trusts the pattern is real -> GWAS needs very large sample sizes because each common variant carries only a small individual effect. 5. Firefighters still have to walk the block by hand and name the exact building before anyone calls it solved -> A flagged gene needs separate functional validation before being accepted as causal.