Skip to main content
. 2014 Nov 1;14:373. doi: 10.1186/s12884-014-0373-3

Figure 1.

Figure 1

Clustering of pregnant versus referent women. Multivariate PLS modelling with the pregnant and referent status as dependent variables and all 66 FFQ foods/food aggregates, nutrients, supplements, and use of tobacco as the block of independent variables. Clustering of participants is displayed in a score loading plot with the two strongest components t[1] and t[2]) on the x- and y-axis for (a) 30 year olds and (b) 40 year olds.