Skip to main content
Figure 1 | BMC Bioinformatics

Figure 1

From: SPiCE: a web-based tool for sequence-based protein classification and exploration

Figure 1

Protein classification. A) Feature extraction maps protein sequences to feature space. In this case, calculation of the sequence length (x-axis) and the relative frequency of occurrence of alanines (y-axis) map each protein sequence to a point in two-dimensional feature space. B) Classifier training using proteins with known class labels: class 1 (orange) and class 2 (green). After mapping to feature space, a classifier is trained to obtain a decision boundary (dashed line) that optimally separates the classes. C) Predicting class labels of new proteins using the trained classifier. After mapping to feature space, the point in feature space determines what label is assigned to the protein. Label class 1 will be assigned to the example protein, because of its location on the class 1 side of the decision boundary.

Back to article page