It should, however, be noted that, given the relatively wide range of exposures represented in this data set, the estimate of variability may be less uncertain than the central tendency estimate.
We pooled studies for each drug using the DerSimonian-Laird random effects model [ 16], which recognizes and anchors studies as a sample of all potential studies, and incorporates an additional between-study component to the estimate of variability [ 17].
For small experiments such as the one we conducted, the estimate of variability for a single test is based on only a few measurements and is therefore not very reliable.
Bootstrap resampling of the data set revealed that estimates of variability and power were associated with considerable uncertainty.
In general, the estimates of genetic variability were moderate.
All the estimates of sample variability are given in terms of the standard error (SE).
This analysis complements the estimates of genetic variability and differentiation presented in Hoeck et al [ 28].
The estimates of the variability in batch, B, and in interaction between array and dye, AD, were small.
Second, our results show that the estimates of baseline variability and between-study heterogeneity remain biased regardless of the number of studies in the meta-analysis.
