Skip to main content
Fig. 4 | BMC Bioinformatics

Fig. 4

From: Utilizing knowledge base of amino acids structural neighborhoods to predict protein-protein interaction sites

Fig. 4

The relationship between prediction quality, threshold and number of most similar elements. The dependence of the prediction quality based on the number of most similar elements used for the prediction (individual lines) and on the threshold (X-axis). The threshold specifies the minimum portion of retrieved elements to be labeled as interface in order to denote the evaluated amino acid as an interface one. (Neighborhood: c 12; fingerprints length: 1023 bits; features type: fingerprints with amino acid type and both amino acid type and RASA value on the central residue)

Back to article page