Skip to main content

Table 7 Comparison of all baseline approaches and ELECTRA-DTA on the Davis datasets

From: ELECTRA-DTA: a new compound-protein binding affinity prediction model based on the contextualized sequence encoding

Dataset Model CI MSE R \(r_m^2\) AUPR
Original Davis Dataset KronRLS 0.871 0.379 0.407 0.661
SimBoost 0.872 0.282 0.644 0.709
DeepDTA 0.876 (0.004) 0.261 0.846 0.630 (0.017) 0.714 (0.010)
WideDTA 0.886 0.262
DeepCDA 0.891 (0.003) 0.248 0.857 0.649 (0.009) 0.739 (0.006)
ELECTRA-DTA 0.897 (0.003) 0.238 0.844 0.671 (0.032) 0.698 (0.010)
refined Davis Dataset DeepDTA 0.882 (0.016) 0.191 0.843 0.690 (0.035) 0.695 (0.03)
Attention-DTA 0.888 (0.007) 0.195 0.836 0.697 (0.005) 0.677 (0.022)
ELECTRA-DTA 0.896 (0.002) 0.195 0.838 0.637 (0.048) 0.685 (0.026)
  1. Bold values represent the best performance over all competitive methods