Skip to main content
Fig. 2 | Genome Biology

Fig. 2

From: Trans-ancestral genome-wide association study of longitudinal pubertal height growth and shared heritability with adult health outcomes

Fig. 2

Six phenotypes assessed by GWAS. The top panel shows a typical growth curve for boys (blue) and girls (red), with age (years) on the x-axis and height gain (cm/year) on the y-axis. First, we included three simple height or height-difference phenotypes as previously assessed [14]: (I) The take-off phase of the pubertal growth spurt (height at age 10 in girls and 12 in boys); (II) total pubertal growth, between ages 8 and adult; and (III) late pubertal growth, between ages 14 and adult. Additionally, we included three phenotypes derived from SITAR longitudinal modeling: (IV) a-size; (V) b-timing; and (VI) c-intensity. The black line represents the mean population growth curve for a cohort (by sex). Each individual gets a random effect for the three parameters; for example, if a subject is taller than their peers (upper red line), they get a positive value for a-size, while a shorter individual (lower line in panel IV) gets a negative value. A subject who enters their growth spurt earlier than the mean (left line in panel V) gets a negative value for b-timing, while a subject growing later (right line in panel V) gets a positive value. Finally, a subject who grows faster than the population mean (steeper line in panel VI) receives a positive value for c-intensity, while a subject growing slower (shallower line in panel VI) gets a negative value for c-intensity

Back to article page