Daniel Crusius
@dan-cr.bsky.social
(Bio-)chemistry + computers @ University of Oxford | AI + Cheminformatics @ PangeaBio | https://danie.lc
Great work! Correlation between training set similarity and prediction accuracy is observed in traditional QSAR modelling. Interesting to see that this seems to also apply when predicting protein-ligand interactions.
doi.org/10.1021/ci04...
doi.org/10.1021/ci04...
Similarity to Molecules in the Training Set Is a Good Discriminator for Prediction Accuracy in QSAR
How well can a QSAR model predict the activity of a molecule not in the training set used to create the model? A set of retrospective cross-validation experiments using 20 diverse in-house activity se...
doi.org
February 8, 2025 at 6:45 PM
Great work! Correlation between training set similarity and prediction accuracy is observed in traditional QSAR modelling. Interesting to see that this seems to also apply when predicting protein-ligand interactions.
doi.org/10.1021/ci04...
doi.org/10.1021/ci04...