We describe, to our knowledge, the first genetic association analyses in UK Biobank, targeting the genetic architecture of smoking behaviour and lung function phenotypes. By sampling from the extremes of the FEV1 and smoking phenotype distributions, we identified novel associations for FEV1 and smoking behaviour. We show genome-wide evidence for shared genetic causes of low FEV1 between heavy smokers and never smokers. Furthermore, our analyses suggest that smoking is only likely to interact with a small proportion of the genetic effects we have identified on lung function—that is, smoking and genetic effects generally act separately. We also show shared genetic causes of airflow obstruction between participants who reported doctor-diagnosed asthma and those who did not.
