TorchCraft framework uses AlphaFold 3 to design molecular binders

Researchers have developed TorchCraft, a new framework for designing molecular binders by inverting an all-atom structure predictor. This method leverages pretrained weights from AlphaFold 3 and is implemented in TorchFold, combining various objectives like confidence, contact, and geometric priors. TorchCraft has demonstrated success in generating minibinders and VHHs that exhibit experimentally verified binding across multiple targets, and it shows potential for designing cyclic peptides and ligand-binding proteins. AI

影响 This framework could accelerate drug discovery and protein engineering by enabling more efficient and accurate design of molecular binders.

排序理由 The cluster describes a new scientific paper detailing a novel computational framework for molecular design. [lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

• AlphaFold 3
• TorchCraft
• TorchFold

• 论文
• 模型发布

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →