Misleading Westerns: Common Quantification Mistakes in Western Blot Densitometry and Proposed Corrective Measures

Figure 3

Assessment of errors in normalisation of Western blot densitometry data. Western blot O.D. data that were poor fits to proportional linear models were normalised to O.D. data from commonly used loading controls. Pregnant human myometrial tissue homogenates extracted in 2D lysis buffer were run in independent dilutions at 40, 20, 10, 5, and 2.5 μg of total protein lysate. A myometrial sample from a woman not-in-labour and a sample from a woman in labour were run on each membrane and this was repeated five times with samples from different women used in each repeat. Densitometry was performed by selecting the highest non-detector saturated exposure that was available. These data were normalised to the indicated loading control from the same lane in each membrane. (a) RhoA normalised to αSMA, (b) MYPT1 normalised to αSMA, (c) RhoA normalised to MYPT1, (d) RhoA normalised to Ponceau S, (e) MYPT1 normalised to Ponceau S, and (f) αSMA normalised to Ponceau S.

(a)

(b)

(c)

(d)

(e)

(f)