Abstract
Densitometry data generated for Western blots are commonly used to compare protein abundance between samples. In the last decade, it has become apparent that assumptions underpinning these comparisons are often violated in studies reporting Western blot data in the literature. These violations can lead to erroneous interpretations of data and may contribute to poor reproducibility of research. We assessed the reliability of Western blot data obtained to study human myometrial tissue proteins. We ran dilution series of protein lysates to explore the linearity of densitometry data. Proteins analysed included αSMA, HSP27, ERK1/2, and GAPDH. While ideal densitometry data are directly proportional to protein abundance, our data confirm that densitometry data often deviate from this ideal, in which case they can fit nonproportional linear or hyperbolic mathematical models and can reach saturation. Nonlinear densitometry data were observed when Western blots were detected using infrared fluorescence or chemiluminescence, and under different SDS-PAGE conditions. We confirm that ghosting artefacts associated with overabundance of proteins of interest in Western blots can skew findings. We also confirm that when data to be normalised are not directly proportional to protein abundance, it is a mistake to use the normalisation technique of dividing densitometry data from the protein-of-interest with densitometry data from loading control protein(s), as this can cause the normalised data to be unusable for making comparisons. Using spiked proteins in a way that allowed us to control the total protein amount per lane, while only changing the amount of spiked proteins, we confirm that nonlinearity and saturation of densitometry data, and errors introduced from normalisation processes, can occur in routine assays that compare equal amounts of lysate. These findings apply to all Western blot studies, and we highlight quality control checks that should be performed to make Western blot data more quantitative.