It is discussed that specific amino acids (AAs) have functional roles in early life. Understanding the AA composition in human milk (HM) during lactation assists in specifying these roles. To this end we assessed the levels of free AAs (FAAs), total AAs (free and bound, TAAs) and protein levels in HM in the first 6 months of lactation, and evaluated possible associations with infant gender. HM samples of 25 healthy Dutch mothers participating in the PreventCD study were collected monthly during the first 6 months of lactation. Of the participating mothers, 12 gave birth to a boy and 13 gave birth to a girl. Analyses of the HM samples revealed that levels of free glutamate, glutamine, aspartate, glycine, and serine significantly increased during months 1–3 of lactation, both in absolute sense and relative to TAA levels. Evaluation of gender differences by mixed model analyses revealed an association between female infant gender and higher protein content (p = 0.0465) and TAA content (p = 0.0362) in HM during the first 3 months of lactation. Furthermore, there was a tendency for an association of male infant gender with higher levels of free glutamine (p = 0.0948) in HM during the first 3 months of lactation. These results show that FAA, TAA and protein levels in HM display a time-specific occurrence during lactation. Moreover, although confirmation is necessary in view of the small sample size, this study indicates that the AA composition in HM shows differential effects of the infant’s sex.

When data are affected by multicollinearity in the linear regression framework, then concurvity will be present in fitting a generalized additive model (GAM). The term concurvity describes nonlinear dependencies among the predictor variables. As collinearity results in inflated variance of the estimated regression coefficients in the linear regression model, the result of the presence of concurvity leads to instability of the estimated coefficients in GAMs. Even if the backfitting algorithm will always converge to a solution, in case of concurvity the final solution of the backfitting procedure in fitting a GAM is influenced by the starting functions. While exact concurvity is highly unlikely, approximate concurvity, the analogue of multicollinearity, is of practical concern as it can lead to upwardly biased estimates of the parameters and to underestimation of their standard errors, increasing the risk of committing type I error. We compare the existing approaches to detect concurvity, pointing out their advantages and drawbacks, using simulated and real data sets. As a result, this paper will provide a general criterion to detect concurvity in nonlinear and non parametric regression models.