Skip to main content
Fig. 7 | BMC Bioinformatics

Fig. 7

From: Risk-conscious correction of batch effects: maximising information extraction from high-throughput genomic datasets

Fig. 7

The plot above demonstrates how the confidence (%) for there being no batch effects reduces as the batch means are compressed proportionately towards zero. The three curves correspond to the first three principal components of Dataset 1. For each point on a given curve, batch means are multiplied by the corresponding value of k for the purposes of computing the confidence value. The diamond markers on the curves denote the k values (i.e. the resulting batch mean compression) for a chosen confidence limit of 95 %

Back to article page