Fig. 6.

Determination of Classification Accuracy Ranges for φ and ψ Torsion Angles. Orange Classification Trees use decision-based machine learning to identify the minimum 95%, 90%, 85%, and 80% accuracy torsion angles classifiers. Classification trees determine accuracy based on prediction of inactive and active states for the respective φ/ψ angle. Classifiers shown do not include branched classifiers. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)