Figure 1
From: Transmembrane helix prediction using amino acid property features and latent semantic analysis
Classification of protein feature vectors of the completely-membrane or completely-nonmembrane type: Figure shows the data points of the training set, and linear classifier learnt from this data. The first two dimensions of the features after principal component analysis are shown in the scattergram. It may be seen that even a simple linear classifier can separate out a large fraction of the data points into the correct class.