Pearl: A Foundation Model for Placing Every Atom in the Right Location

TL;DR

Pearl is a new foundation model for protein-ligand structure prediction that leverages synthetic data, SO(3)-equivariant architectures, and controllable inference to achieve state-of-the-art accuracy and generalization in drug discovery applications.

Contribution

The paper introduces Pearl, a novel model combining synthetic data, equivariant diffusion architecture, and flexible inference for improved protein-ligand cofolding predictions.

Findings

Outperforms AlphaFold 3 and baselines on public benchmarks with 14%+ improvements.

Achieves 3.6x better results on challenging real-world drug targets.

Model performance scales with the size of synthetic training data.

Abstract

Accurately predicting the three-dimensional structures of protein-ligand complexes remains a fundamental challenge in computational drug discovery that limits the pace and success of therapeutic design. Deep learning methods have recently shown strong potential as structural prediction tools, achieving promising accuracy across diverse biomolecular systems. However, their performance and utility are constrained by scarce experimental data, inefficient architectures, physically invalid poses, and the limited ability to exploit auxiliary information available at inference. To address these issues, we introduce Pearl (Placing Every Atom in the Right Location), a foundation model for protein-ligand cofolding at scale. Pearl addresses these challenges with three key innovations: (1) training recipes that include large-scale synthetic data to overcome data scarcity; (2) architectures that incorporate an SO(3)-equivariant diffusion module to inherently respect 3D rotational symmetries, improving generalization and sample efficiency, and (3) controllable inference, including a generalized multi-chain templating system supporting both protein and non-polymeric components as well as dual unconditional/conditional modes. Pearl establishes a new state-of-the-art performance in protein-ligand cofolding. On the key metric of generating accurate (RMSD < 2 \r{A}) and physically valid poses, Pearl surpasses AlphaFold 3 and other open source baselines on the public Runs N' Poses and PoseBusters benchmarks, delivering 14.5% and 14.2% improvements, respectively, over the next best model. In the pocket-conditional cofolding regime, Pearl delivers $3.6\times$ improvement on a proprietary set of challenging, real-world drug targets at the more rigorous RMSD < 1 \r{A} threshold. Finally, we demonstrate that model performance correlates directly with synthetic dataset size used in training.