ArticleNature machine intelligence2025
Resolving data bias improves generalization in binding affinity prediction.
Article in Nature machine intelligence, 2025. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 19 papers.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
19 citing papers in PubMed.
- Physical priors improve performance of structure-based binding affinity models.bioRxiv : the preprint server for biology · 2026Article
- Artificial intelligence-assisted lead optimization in drug discovery: bridging computational advances and translational challenges.Medicinal chemistry research : an international journal for rapid communications on design and mechanisms of action of biologically active agents · 2026Review
- A Site-Aware Representation Learning Framework For Unified Molecular Interaction Modeling and Generative Design.Advanced science (Weinheim, Baden-Wurttemberg, Germany) · 2026Article
- Next-generation probiotics: an outlook into current applications and future developments.Nature reviews. Microbiology · 2026Review
- PLATE-VS: a web server for protein-ligand assay curation and cross-target virtual screening datasets.Nucleic acids research · 2026Article
- Artificial intelligence and automation in enzyme engineering: evolution, advances, and future perspectives.Bioresources and bioprocessing · 2026Review
- Structure-aware artificial intelligence for next-generation drug discovery: from protein-ligand modeling to generative biomolecular design.Briefings in bioinformatics · 2026Review
- Reliable evaluation and learning in multi-input biological association prediction.Briefings in bioinformatics · 2026Article
- Absolute Binding Free Energy Calculations between the SARS-CoV-2 Main Protease and 130 Drug Leads Using Implicit Ligand Theory.Journal of chemical information and modeling · 2026Article
- More protein-ligand data are needed for AlphaFold-like models to enable drug discovery.Current opinion in structural biology · 2026Review
- GRIPHIN: grids of pharmacophore interaction fields for affinity prediction.Journal of cheminformatics · 2026Article
- Contemporary data-driven innovations in peptide-based therapeutic design.Briefings in bioinformatics · 2026Review
- From Algorithms to Assets: A Comprehensive Review of AI's Role in Preclinical Drug Discovery and the Hurdles to Clinical Translation.Pharmaceuticals (Basel, Switzerland) · 2026Review
- BA-Pred and RMSD-Pred: Integrated Graph Neural Network Models for Accurate Protein-Ligand Binding Affinity and Binding Pose Prediction.Journal of chemical information and modeling · 2026Article
- Blind Challenges Let Us See the Path Forward for Predictive Models.Journal of chemical information and modeling · 2026Article
- Leak Proof PDBBind: A Reorganized Data Set of Protein-Ligand Complexes for More Generalizable Binding Affinity Prediction.The journal of physical chemistry. B · 2026Article
- Structure-informed machine learning for drug discovery: a task-centric perspective.Briefings in bioinformatics · 2026Review
- Beyond Random Splits: Assessing the Generalization of Graph and Vector Models for WT-Structure-Only Drug Resistance Prediction under Protein-Disjoint Evaluation.Computational and structural biotechnology journal · 2026Article
- Architectural good practices for reproducible benchmarking in protein machine learning.Frontiers in bioinformatics · 2026Article
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
6 authors.
Funding
No grant is acknowledged in the PubMed record.
Abstract
The field of computational drug design requires accurate scoring functions to predict binding affinities for protein-ligand interactions. However, train-test data leakage between the PDBbind database and the Comparative Assessment of Scoring Function benchmark datasets has severely inflated the performance metrics of currently available deep-learning-based binding affinity prediction models, leading to overestimation of their generalization capabilities. Here we address this issue by proposing PDBbind CleanSplit, a training dataset curated by a new structure-based filtering algorithm that eliminates train-test data leakage as well as redundancies within the training set. Retraining current top-performing models on CleanSplit caused their benchmark performance to drop substantially, indicating that the performance of existing models is largely driven by data leakage. By contrast, our graph neural network model maintains high benchmark performance when trained on CleanSplit. Leveraging a sparse graph modelling of protein-ligand interactions and transfer learning from language models, our model is able to generalize to strictly independent test datasets.
Indexed as
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.