In conclusion, we observed an optimum in IGF-I levels and final body height for the 192-bp and 194-bp allele of the IGF-I gene. A gender-specific effect of the IGF-I alleles on body height was observed. The secular trend in body height observed in our elderly Dutch population was similar for the different genotypes; carriers of the 192-bp and/or the 194-bp allele remained significantly taller throughout time.

With a rapid increase in volume and complexity of data sets, there is a need for methods that can extract useful information, for example the relationship between two data sets measured for the same persons. The Partial Least Squares (PLS) method can be used for this dimension reduction task. Within life sciences, results across studies are compared and combined. Therefore, parameters need to be identifiable, which is not the case for PLS. In addition, PLS is an algorithm, while epidemiological study designs are often outcome-dependent and methods to analyze such data require a probabilistic formulation. Moreover, a probabilistic model provides a statistical framework for inference. To address these issues, we develop Probabilistic PLS (PPLS). We derive maximum likelihood estimators that satisfy the identifiability conditions by using an EM algorithm with a constrained optimization in the M step. We show that the PPLS parameters are identifiable up to sign. A simulation study is conducted to study the performance of PPLS compared to existing methods. The PPLS estimates performed well in various scenarios, even in high dimensions. Most notably, the estimates seem to be robust against departures from normality. To illustrate our method, we applied it to IgG glycan data from two cohorts. Our PPLS model provided insight as well as interpretable results across the two cohorts.

Summary. Factor VIII activity (factor VIII:C) and factor VIII antigen (factor VIII:Ag) levels above150 IU/dl are associated with a five-to sixfold increased risk of venous thrombosis compared with levels , 100 IU/dl. These high levels are present in 25% of patients with a first episode of deep-vein thrombosis and in 11% of healthy controls. von Willebrand factor (VWF) and blood group are important determinants of the factor VIII level in plasma and therefore contribute to thrombotic risk, while factor VIII appears to be the final effector. Previously, we found familial clustering of factor VIII:C levels in women, which remained after adjustment for VWF and blood group. In the present study, we analysed the familial influence on factor VIII:Ag levels exceeding 150 IU/dl in 12 large families with thrombophilia in which high factor VIII:Ag levels contribute to thrombotic risk. As expected, blood group was a main determinant of the plasma factor VIII level: 58 relatives (32%) had factor VIII levels above 150 IU/dl and 50 (86%) of these had blood group non-O. After adjustment for blood group and age, we found an association between factor VIII:Ag levels in sister pairs (0´35, P 0´003), brother pairs (0´35, P 0´003), brother±sister pairs (0´35, P , 0´001) and in mother±son pairs (0´45, P 0´02), but not in father±daughter or father±son pairs. The familial aggregation test was strongly positive for factor VIII:Ag levels (P , 0´001) and remained so after adjustment for the influence of blood group. We conclude that high factor VIII:Ag levels are a highly prevalent risk factor for venous thrombosis and contribute to risk in families with thrombophilia, and that these high levels are likely to be genetically determined by factors other than just blood group.