Figure 2

Percentage coefficient of variation by number of sites collected on normal and preeclamptic placentae. To determine the minimum number of sites on a placenta that must be sampled to achieve a coefficient of variation (CV) below 10% for each transcript, all possible combinations of sites had to be considered. To achieve this, we developed a web-based algorithm that took every possible set of 2 samples (66 in total) and calculated the mean of each (http://iconvm.com.au/jonomics). The %CV was then calculated for the 66 means generated. This was repeated using every possible set of 3 samples (220 combinations in total) and so on up to 12 samples, each time calculating the %CV for the set of means generated. This figure is a plot of %CV against the number of sites. The minimum number of sites required to obtain a %CV of less than 10% for both normal and preeclamptic placenta is shown for (a) mFLT-1, (b) sFLT-1 i13 and (c) sFLT-1 e15a.