… equally as bad (or even worse) on the validation set (PCC 0.51, … expected, we aim at using
po-sco for scoring of docked poses, … using maximum likelihood estimation. In addition to the …

… and dynamics information of protein-ligand interactions at the … top scoring docked output for
each protein-ligand complex. … benchmark MD dataset for machine-learning training that can …

… ID-Score [40] predicts protein–ligand binding affinity through … the negative protein–ligand
complexes that either interact with … construct the protein and protein–ligand interaction graphs. …

… models with different adjustable parameters and each showed … trained with both positive
data augmentation and negative … To conduct a comparative assessment of PIGNet2 with other …

… and regenerate parameters for a specific FF using the drop-… in discriminating bad
protein–ligand interactions from the good … to docking scoring. In addition, we confirmed these …

… is considered as negative for the input training datasets. The … the one with high validation
score is considered as an inhibitor … other performance parameters once the training has been …

… All affinities were expressed by a negative log scale. … improved the performance of AK-Score,
we therefore integrated the … atoms not critical for protein–ligand interactions are considered …

… to protein–ligand complexes and added an angular parameter … very unlikely to interact with
other groups of negative (partial) … for an exploration of its use for scoring and ligand design. A …

… to estimate the protein–ligand interaction scores [89]. A sample … scoring functions, these
methods also need to be trained on … However, more recently “negative” data have also been …

… the positive and negative example pairs into training (n = 2, … Fixing the remaining parameters,
we perform global search … pocket pairs from the training set of the scoring neural network (…