Skip to main content

Table 9 Results of ablation experiments

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

Dataset

method

CI

MSE

R

\(r_m^2\)

AUPR

Davis

ELECTRA-DTA

0.896 (0.002)

0.195

0.838

0.637 (0.048)

0.685 (0.026)

Onehot-DTA

0.850 (0.005)

0.301

0.739

0.525 (0.018)

0.561 (0.029)

KIBA

ELECTRA-DTA

0.892 (0.002)

0.143

0.892

0.780 (0.014)

0.805 (0.005)

Onehot-DTA

0.883 (0.002)

0.157

0.881

0.744 (0.012)

0.792 (0.003)

BindingDB

ELECTRA-DTA

0.837 (0.002)

0.650

0.860

0.670 (0.027)

0.811 (0.001)

Onehot-DTA

0.830 (0.001)

0.700

0.849

0.659 (0.037)

0.799 (0.004)

  1. Bold values represent the best performance over all competitive methods