Abstract
The proximity extension assay (PEA) enables large-scale proteomic investigations across numerous proteins and samples. However, discrepancies between measurements, known as batch-effects, potentially skew downstream statistical analyses and increase the risks of false discoveries. While implementing bridging controls (BCs) on each plate has been proposed to mitigate these effects, a clear method for utilizing this strategy remains elusive. Here, we characterized batch effects in PEA proteomics and identified three types: protein-specific, sample-specific, and plate-wide. We developed a robust regression-based method called BAMBOO (Batch Adjustments using Bridging cOntrOls) to correct them. Simulations comparing BAMBOO with established correction techniques (median centering, median of the difference (MOD), and ComBat) revealed that median centering and ComBat were significantly impacted by outliers within the BCs, whereas BAMBOO and MOD were more robust when no plate-wide effects were introduced. Optimal batch correction was achieved with 10-12 BCs. We validated the simulation results using experimental data and found that BAMBOO and MOD had a reduced incidence of false discoveries compared to alternative methods. Our findings emphasize the prevalence of batch effects in PEA proteomic studies and advocate for BAMBOO as a robust and effective tool to enhance the reliability of large-scale analyses in the proteomic field.