跳到正文
Molence· @MolenceBio · X·· 17 天前AI 评分40
AI 导读

11/15 💊 在陌生结合环境中实现更强的配体预测。 在 Runs N’ Poses 上,Mol-1 在全部八个相似度区间内领先所有被评估基线。 在相似度最低的 0–20 区间,成功率为 36.1%,AlphaFold 3 为 24.6%。通过 Mol-1-constraint 提供结合位点信息后,成功率提升至 63.9%。 成功需同时满足配体 RMSD < 2 Å 和 lDDT-PLI > 0.8。 #DrugDiscovery

正文 · 原文

11/15 💊 Stronger ligand prediction in unfamiliar binding environments.

On Runs N’ Poses, Mol-1 leads the evaluated baselines across all eight similarity bins.

In the lowest-similarity 0–20 bin, success is 36.1%, versus 24.6% for AlphaFold 3. Supplying binding-site information through Mol-1-constraint raises it to 63.9%.

Success requires both ligand RMSD < 2 Å and lDDT-PLI > 0.8.

#DrugDiscovery

引用Molence@MolenceBio
10/15 🎯 Designing where a VHH binds. For TNFR2 and IL-4Rα, we targeted the natural ligand-binding regions. In sequential BLI competition experiments, prebinding TNFα or IL-4 reduced subsequent binding by the corresponding designed VHH. These experiments support recognition of the intended regions, connecting an epitope-directed design objective to experimental evidence. #ProteinDesign
在 X 查看被引用的帖子

来源:Molence · x.com