Using a cross‐model loadings plot to identify protein spots causing 2‐DE gels to become outliers in PCA

Abstract

The multivariate method PCA is an exploratory tool often used to get an overview of multivariate data, such as the quantified spot volumes of digitized 2-DE gels. PCA can reveal hidden structures present in the data, and thus enables identification of potential outliers and clustering. Based on PCA, we here present an approach for identification of protein spots causing 2-DE gels to become outliers. The approach can potentially obviate analytical exclusion of entire 2-DE gels.