Skip to main content

Table 4 The detailed statistics of the refined datasets, containing the number of proteins, compounds, interactions and average number of interactions per protein and per compound

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

  No. Proteins No. Compounds No. interaction No. interactions per protein No. interactions per compound No. inactive interactions No. non-inactive interactions
Davis 361 68 24548 68 61 1649 22899
KIBA 229 2052 117184 511.720 57.107 24543 92641
BindingDB 1615 129109 144525 79.944 1.541 41487 87622