Binding

Aug 25, 2026

Accounting for the unbound state improves inverse-folding prediction of binding-affinity changes

Aug 25, 2026

AlphaFold3 ipTM can distinguish between antibody binders and nonbinders

Aug 25, 2026

Boltz-2 intermediate representations can predict protein-protein binding affinity

Aug 25, 2026

The Boltz-2 affinity module cannot be effectively repurposed for PPI affinity prediction

Jul 28, 2026

AF3 binding affinity predictions are orthogonal to those made by force fields and other neural networks

Jul 28, 2026

All-atom structure and affinity predictors partially generalize to point mutants

Jul 28, 2026

Large-scale measurements, such as Kd and IC50, from different assays correlate weakly in different studies

Jul 20, 2026

Explicit germline encoding improves zero-shot antibody-affinity prediction

May 06, 2026

ESM-IF outperforms ProteinMPNN on Ab inverse folding and affinity prediction

Apr 21, 2026

Generic PLMs outperform antibody-specific PLMs on zero-shot predictions of affinity changes

Apr 21, 2026

PAE weakly correlates with Ab-Ag binding

Apr 21, 2026

Antibody PLMs match or outperform generic PLMs on specificity prediction

Apr 21, 2026

Attention matrices from antibody LMs can be used for paratope prediction

Apr 21, 2026

Classifiers can be used as predictors of continuous properties

prediction/binding

Apr 21, 2026

Composition of negative data affects rule discovery in binary classification tasks

Apr 21, 2026

Deep learning methods cannot generalize T-cell receptor binding to new epitopes

Apr 21, 2026

ESM-IF representations can predict if residue binds ligands

Apr 21, 2026

Free energy perturbation

Apr 21, 2026

Insufficient ddG data on Abs for training

Apr 21, 2026

Monomeric AF2 is better able to distinguish positive and negative PPIs than AF2-multimer

Apr 21, 2026

PLM embeddings fine-tuned using contrastive learning outperform other representations in drug-target interaction prediction

Apr 21, 2026

Protein language models can predict zero-shot which proteins belong to the same species

Apr 21, 2026

Structure-based methods outperform sequence-based methods at zero-shot prediction of binding, whereas the reverse is true for zero-shot prediction of enzymatic activity

Apr 21, 2026

Training protein structure prediction neural networks on both positive and negative protein-protein interactions improves PPI discrimination

Apr 21, 2026

The Evoformer can calculate ligand-binding residues

Apr 20, 2026

Quartz 4

Explorer

Binding

Accounting for the unbound state improves inverse-folding prediction of binding-affinity changes

AlphaFold3 ipTM can distinguish between antibody binders and nonbinders

Boltz-2 intermediate representations can predict protein-protein binding affinity

The Boltz-2 affinity module cannot be effectively repurposed for PPI affinity prediction

AF3 binding affinity predictions are orthogonal to those made by force fields and other neural networks

All-atom structure and affinity predictors partially generalize to point mutants

Large-scale measurements, such as Kd and IC50, from different assays correlate weakly in different studies

Explicit germline encoding improves zero-shot antibody-affinity prediction

ESM-IF outperforms ProteinMPNN on Ab inverse folding and affinity prediction

Generic PLMs outperform antibody-specific PLMs on zero-shot predictions of affinity changes

PAE weakly correlates with Ab-Ag binding

Antibody PLMs match or outperform generic PLMs on specificity prediction

Attention matrices from antibody LMs can be used for paratope prediction

Classifiers can be used as predictors of continuous properties

Composition of negative data affects rule discovery in binary classification tasks

Deep learning methods cannot generalize T-cell receptor binding to new epitopes

ESM-IF representations can predict if residue binds ligands

Free energy perturbation

Insufficient ddG data on Abs for training

Monomeric AF2 is better able to distinguish positive and negative PPIs than AF2-multimer

PLM embeddings fine-tuned using contrastive learning outperform other representations in drug-target interaction prediction

Protein language models can predict zero-shot which proteins belong to the same species

Structure-based methods outperform sequence-based methods at zero-shot prediction of binding, whereas the reverse is true for zero-shot prediction of enzymatic activity

Training protein structure prediction neural networks on both positive and negative protein-protein interactions improves PPI discrimination

The Evoformer can calculate ligand-binding residues

Kd differs from IC50, LD50, and GI50