Vilya-1 Improves Macrocycle Conformation Prediction
DaveJuergens · x · 2026-07-18
Vilya-1 is an all-atom diffusion-based foundation model designed for macrocycle conformation prediction and design. It aims to solve a core bottleneck in drug discovery: stably sampling biologically relevant, low-energy conformations across diverse synthesizable chemical spaces, moving beyond standard peptides.
The authors report that across 66 cyclic peptide X-ray structures, Vilya-1 achieved an 89.2% success rate in generating near-native ring conformations (ring RMSD < 1 Å), significantly outperforming Prime-MCS, RDKit ETKDGv3, Boltz-2/RF3, and several deep learning conformational generators.
The model also generalizes across various macrocycle types and topologies, including:
- Disulfide-linked cyclic peptides
- Sidechain-to-sidechain cyclization
- Tail-to-sidechain cyclization
- Non-peptidic macrocycles, such as macrolides/polyketides
The post also mentions FK506 among the example molecules.
More from Research
- NLI alone rejects 5/6 new agent memories; a 0.45 similarity gate fixes it all — Alternative_Pin9598 · 2026-10-26
- AI folks, stop saying "biology" when you mean drug development — KevinKaichuang · 2026-10-26
- ostris pre-training "Primal VAE": 1-channel latents match modern 8x VAEs via patching — ostrisai · 2026-10-26
- Shanghai AI Lab-led team releases 100-hour egocentric wearable dataset for humanoid training — ChuckDBrooks · 2026-10-26
- ConvFill: A 0.6B Talker Buys Time While a Large Reasoner Whispers the Real Answer — andimarafioti · 2026-10-26
- Flow Reasoning Models turn flow matching into iterative reasoners, hitting 99.5% on Sudoku-Extreme — alec_helbling · 2026-10-26