Peptide Permeability Measurements in PAMPA: Sources of Variability and Implications for Data Reproducibility

Peptide permeability is increasingly assessed using parallel artificial membrane permeability assays (PAMPA), yet reported permeability values often show large variability and poor cross-study comparability. Here, we systematically investigate key experimental and analytical factors that influence permeability measurements by performing controlled, data-driven studies across multiple assay conditions. Using a set of representative peptides and small-molecule controls, we quantitatively assess the effects of incubation time, initial concentration, temperature, evaporation, detection workflow, and well-to-well variability on measured permeability rates. We show that transport flux frequently deviates from the linear assumptions underlying apparent (P app) and effective (P e) permeability calculations, particularly due to lag times, concentrationdependent behavior, and signal loss near detection thresholds. In addition, we demonstrate that commonly under-reported experimental details, such as temperature control, evaporation mitigation, and desalting prior to LC-MS analysis, can introduce 1 systematic bias that is not fully corrected by internal standards. Together, these results reveal that PAMPA permeability values are highly sensitive to both experimental design and data analysis choices, limiting direct comparison across studies unless conditions are carefully controlled and reported. We conclude by outlining practical recommendations for conducting, analyzing, and reporting peptide PAMPA experiments, to improve reproducibility and enable more reliable use of PAMPA data in peptide lead optimization and decision-making workflows.