Fig. 4

Seasonal variation of mitochondrial abundance associated with incident disease risk. We used linear regression to model the impact of the sine wave on incident disease risk in the UK Biobank. We found that the amplitude and displacement but not acrophase were associated with major incident disease groups. Any disease indicates any of the investigate disease while multiple diseases denote the development of two or more major disease groups. # = all/multiple diseases excluding musculoskeletal and autoimmune diseases and neoplasms. * = P < 0.05, ** = P < 0.01, *** = P < 0.001