Overview of GUSTO

The primary hypothesis to be evaluated is that a mix of genomic and epigenomic measures at birth are influenced by prenatal factors and will predict patterns of growth and body composition in infancy and childhood. Mixed models will be employed to evaluate the extent in which these epigenetic measures are influenced by repeated prenatal and childhood growth factors. As infant anthropometry will be measured repeatedly over time, the implementation of mixed model would take into account the possible correlations between these repeated measurements. Correlations between repeated intrauterine ultrasound scans will also be accounted for. Additional phenotypic data such as gender, parity, birth size, developmental data, and infant feeding data may also be considered in the model. All statistical evaluations will be made based on two-sided tests, and multiple comparisons will be accounted for using Bonferroni correction or false discovery rate when a large number of epigenetic factors are evaluated. The commercially available statistical programs SAS and R will be used for analyses. Interim data analyses will be performed yearly to yield preliminary results and check on the current status of the study. Data will be captured using portable computer-assisted technology and held through the systems and processes for confidentiality maintained by the LORIS database system. Initial database entry and computation checks with regular audits will be made to ensure data quality.


1. Rankinen T, Zuberi A, Chagnon YC, et al. The human obesity gene map: the 2005 update.
Obesity (Silver Spring) 2006;14:529-644.

2. Gallaher BW, Breier BH, Keven CL, Harding JE, Gluckman PD. Fetal programming of insulin-like growth factor (IGF)-I and IGF-binding protein-3: evidence for an altered response to undernutrition in late gestation following exposure to periconceptual undernutrition in the sheep. J Endocrinol 1998;159:501-8.

Related Link