using technical replicates for batch bridging of illumina epic methylation data

Question

[jan.haas]

Hey, I have Illumina EPIC methylation data of about 500 samples measured in 9 batches including 25 pairs of technical replicates across different batches. Can someone help how to best use these measurements for bridging between the batches and averaging out the duplicates to have a final matrix with only unique samples?

Thanks!