TorchCraft uses protein prediction to create new molecular binders
TorchCraft: Unified binder design by inverting an all-atom structure predictor
Artificial IntelligenceComputational Engineering, Finance, and ScienceMachine Learning
Summary
Designing molecules that can stick to specific biological targets is difficult, especially when using complex protein structure predictions. The authors developed TorchCraft, a system that inverts a detailed protein structure predictor to design new binding molecules. TorchCraft works with several types of binders, including small proteins and peptides, generating functional candidates without extra redesign steps. It uses the advanced protein modeling abilities of AlphaFold 3 to guide the design process.
What this means in practice
- •For biotech engineers: Create novel protein or peptide binders targeting specific biomolecules using an integrated structure prediction and design framework.$Commercial implications: Provides a platform for designing custom molecular binders that can improve drug discovery pipelines and therapeutic development.
- •For computational chemists: Design cyclic peptides and ligand-binding pockets through joint optimization guided by a frozen all-atom structural model, accelerating molecular design workflows.
Authors
TorchCraft Team, Yu Liu, Zhouhanyu Shen, Zhengyi Li, Xikun Huang, Jiaqi Liu, Shuxian Gao, Qilin Yu, Xiayan Qin, Yucheng Zhang, Mingchen Chen
Abstract
All-atom structure predictors model diverse molecular interactions, but using their learned structural priors for binder design remains challenging. Here we present TorchCraft, a unified binder-design framework that optimizes sequence logits through a frozen all-atom predictor. Implemented in TorchFold, TorchCraft combines confidence, contact, geometric, and sequence-prior objectives within a shared optimization procedure for minibinders, framework-conditioned VHHs, cyclic peptides, and ligand-binding proteins. Using pretrained AlphaFold 3 weights, TorchCraft generated representative minibinders and VHHs with experimentally measured binding across four targets in each format, without post hoc sequence redesign. Computational benchmarks further demonstrated the framework's applicability to cyclic peptides and ligand-conditioned pocket design. TorchCraft extends predictor inversion to multiple binder formats and molecular contexts, providing a common framework for reusing all-atom structural priors in design.