Interpolation-Based Conditioning Of Flow Matching Models For Bioisosteric Ligand Design
Ziv, Y.; Buttenschoen, M.; Scheibelberger, L.; Marsden, B.; Deane, C.
Show abstract
AO_SCPLOWBSTRACTC_SCPLOWFast, unconditional 3D generative models can now produce high-quality molecules, but adapting them for specific design tasks often requires costly retraining. To address this, we introduce two training-free, inference-time conditioning strategies, Interpolate-Integrate and Replacement Guidance, that provide control over E(3)-equivariant flow-matching models. Our methods generate bioisosteric 3D molecules by conditioning on seed ligands or fragment sets to preserve key determinants like shape and pharmacophore patterns, without requiring the original fragment atoms to be present. We demonstrate their effectiveness on three drug-relevant tasks: natural product ligand hopping, bioisosteric fragment merging, and pharmacophore merging.
Matching journals
The top 3 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- FlowPacker: Protein side-chain packing with torsional flow matching 96%
- Cross-Modality and Self-Supervised Protein Embedding for Compound-Protein Affinity and Contact Prediction 95%
- DTI-Voodoo: machine learning over interaction networks and ontology-based background knowledge predicts drug-target interactions 95%
Similar papers in this journal
- DiffDock-Glide: a hybrid physics-based and data-driven approach to molecular docking 96%
- DrugHIVE: Target-specific spatial drug design and optimization with a hierarchical generative model 95%
- DENVIS: scalable and high-throughput virtual screening using graph neural networks with atomic and surface protein pocket features 95%
Similar papers in this journal
- DrugDiff - small molecule diffusion model with flexible guidance towards molecular properties 96%
- Sequence-based Drug-Target Complex Pre-training Enhances Protein-Ligand Binding Process Predictions Tackling Crypticity 96%
- DeepGraphMol, a multi-objective, computational strategy for generating molecules with desirable properties: a graph convolution and reinforcement learning approach 95%
Similar papers in this journal
- SurfDock is a Surface-Informed Diffusion Generative Model for Reliable and Accurate Protein-ligand Complex Prediction 95%
- Atom level enzyme active site scaffolding using RFdiffusion2 95%
- OpenFold: Retraining AlphaFold2 yields new insights into its learning mechanisms and capacity for generalization 93%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.