Figure 5: % of CASP14 (blue) and CASP15 (red) protein
assembly targets with high quality (>0.8) and low quality
(<0.5) computed models as measured by the average contact
precision (left) and recall (right ) (11). In CASP15 the fraction
of high-quality models increased from less than 10% to more than 60%
by both measures. Correspondingly, the fraction of poor-quality models
dropped precipitously.
As figure 5 shows, the improvement in performance compared to the
previous CASP is dramatic, and attributable to deep learning methods
displacing earlier classical docking and homology methods.