Blitz is a few‑step distillation framework that accelerates diffusion‑based protein structure prediction models by reducing denoising steps from 200 to eight. It employs a geometric noise schedule, multistep consistency training, physical steering, and target‑free candidate selection to maintain high structural accuracy. Applied to AlphaFold 3 and Boltz‑2, Blitz matches teacher‑level accuracy and outperforms other few‑step baselines across multiple evaluation metrics.
arXiv:2606. 08375v1 Announce Type: new Abstract: All-atom generative modeling of 3D biomolecular complexes has emerged as the dominant paradigm for predicting the structure of proteins and protein-ligand systems.
MLIPDiff is an inference‑time guidance technique that enhances pretrained denoising diffusion models for ligand generation by incorporating physics‑informed guidance from a machine‑learned interatomic potential (UMA). It combines four weighted energy terms—protein‑ligand interaction, ligand strain, steric clash, and pocket anchoring—to eliminate protein‑ligand collisions while preserving local geometry, without requiring model retraining. On the CrossDocked2020 benchmark, MLIPDiff produces clash‑free, physically plausible ligand poses and achieves superior binding‑affinity metrics compared to both the base diffusion model and NucleusDiff.
The paper introduces two closed‑form projection operators that enforce physical validity in biomolecular diffusion models during inference. One operator resolves inter‑chain van der Waals clashes, while the other restores ligand geometry, including bond lengths, angles, planarity, and chirality. These lightweight, model‑agnostic corrections preserve structural accuracy and ligand placement across multiple benchmarks without adding runtime or memory overhead.
CryoAtlas is a large, rigorously curated dataset and unified benchmark for density‑guided atomic model building of protein chains from cryo‑EM maps. It compiles 17,809 map‑model‑sequence entries from EMDB and PDB (up to March 2026), with a 16,657‑entry training split, and defines a 618‑target benchmark that evaluates structural accuracy, map fit, and stereochemical validity. The resource enables systematic comparison of learning‑based methods with traditional tools like Phenix and AlphaFold3, and demonstrates that retraining a CNN component on CryoAtlas improves performance.
arXiv:2606. 19377v1 Announce Type: cross Abstract: Computational enzyme design requires generating proteins that scaffold catalytic residues and ligands, a task that demands both geometric accuracy and structural diversity from the underlying generative model.
arXiv:2608. 11475v1 Announce Type: cross Abstract: AlphaFold3-class structure predictors pair a representational trunk, which processes sequence and context, with a diffusion module, which generates atomic coordinates.
AtomSteer is a test‑time inference framework that steers pretrained atomistic generative models to satisfy structural constraints such as crystallographic symmetry, geometry, and stereochemistry. It applies bounded local projections, graph‑rigid stereochemical transformations, and periodic symmetry projections during sampling, accepting updates that improve constraint satisfaction while preserving the rest of the structure. The method is demonstrated on three backbones—FlowMM, ET‑Flow, and Boltz‑2—across tasks including de novo crystal generation, crystal structure prediction, molecular conformer generation, and protein‑ligand complex prediction, yielding improvements in space‑group coverage, stereochemical accuracy, and ligand E/Z error rates.
arXiv:2602. 24007v3 Announce Type: replace-cross Abstract: Protein function relies on dynamic conformational ensembles, yet current generative models like AlphaFold3 often fail to produce ensembles that match experimental data.
arXiv:2606. 31332v1 Announce Type: new Abstract: Protein automodeling from cryo-EM density maps faces unique challenges in enforcing physicochemical validity and managing conformational heterogeneity.
arXiv:2606. 06717v1 Announce Type: cross Abstract: While generative AI models have demonstrated remarkable success in structure-based drug design, they predominantly rely on deep binding pockets and struggle to sample effective ligands for challenging low-pocketability targets, such as the historically "undruggable" oncology targets KRAS and MYC.
arXiv:2607. 28553v1 Announce Type: new Abstract: Predicting the 3D structures of atomic systems is fundamental to advancing material science and drug discovery.
MAST is a motif‑augmented diffusion framework that uses a search tree to jointly elucidate 2D and 3D molecular structures from spectroscopic data. It incorporates explicit motif priors as intermediate evidence during denoising, reducing spectral ambiguity, and casts diffusion sampling as a reward‑guided tree search to prioritize high‑reward trajectories, thereby limiting computational overhead. On the QM9S multi‑spectra benchmark, MAST achieves 94.89% exact recovery, improves 3D fidelity, and maintains high chemical validity and stability.