Accelerating Biomolecular Modeling with AtomWorks and RF3 🚀 New preprint from David Baker!🚀 1. A new framework called AtomWorks has been introduced to revolutionize biomolecular modeling. AtomWorks provides a unified and modular platform for developing state-of-the-art biomolecular models, including structure prediction, protein design, and sequence design. It streamlines the process of data preparation and model training, making it easier for researchers to prototype and test new ideas. 2. The AtomWorks framework emphasizes high-quality data handling. It standardizes inputs from diverse sources, such as the Protein Data Bank (PDB), and resolves common issues like incorrect bond orders, charges, and missing coordinates. This results in higher-quality derived features and improved model performance. For example, AtomWorks-generated reference conformers have lower energies compared to those from other open-source models. 3. AtomWorks enables rapid prototyping by breaking down data processing and featurization into modular components. This modular design allows researchers to reuse core building blocks across different networks and easily add new features. It also simplifies the integration of various datasets, facilitating the training of models like RF3 on a diverse set of biomolecular structures. 4. The framework supports scalable training of biomolecular models. AtomWorks shares most of its code across different networks, allowing researchers to repurpose existing components and improve common operations. This efficiency is demonstrated by the ability to process large batches of data quickly, such as processing a 10,000-token batch through the LigandMPNN pipeline in the time it takes for a single forward/backward pass. 5. AtomWorks is accompanied by industry-grade testing and comprehensive documentation. This ensures that the framework is reliable and easy to use, even for researchers without extensive software development experience. The documentation includes worked examples illustrating how to develop pipelines for various biomolecular modeling tasks. 6. Using AtomWorks, the authors trained RosettaFold-3 (RF3), an all-atom biomolecular structure prediction network. RF3 incorporates novel features such as implicit chirality representations and atom-level geometric conditioning, which improve its performance on tasks like predicting chiral ligands and fixed-backbone conformations. 7. RF3 simplifies dataset integration by supporting direct loading from raw crystallographic information files (CIF). The authors introduced new distillation datasets, including a nucleic acid complex distillation set and an RNA distillation set, to enhance the model's training. Additionally, RF3 includes a disordered distillation set to address issues with hallucinated secondary structures. 8. RF3 accurately adheres to specified stereochemistry out-of-the-box, without requiring inference-time guidance. It represents stereochemistry by the sign of the angles formed by atoms surrounding each chiral center and uses data augmentation techniques to improve chirality handling. As a result, RF3 predicts the correct chirality for 88% of ligand chiral centers in the test set, compared to 84% for AlphaFold3 and 76% for Boltz-2. 9. RF3 enables flexible user control through arbitrary atom-level conditioning. Users can specify distances between atoms to incorporate experimentally derived constraints, perform protein-ligand docking, or fold proteins around specific ligand conformers. This feature significantly improves the accuracy of protein-ligand interface predictions. 10. RF3 narrows the performance gap between existing open-source structure prediction models and AlphaFold3. It demonstrates competitive performance on various tasks, such as predicting protein-protein interfaces, protein-ligand interactions, and mixed L/D peptides. When trained on data up to January 2024, RF3 shows further improvements in performance. 11. The authors also trained ProteinMPNN and LigandMPNN using AtomWorks, demonstrating comparable performance to the original models. This highlights the versatility of the AtomWorks framework in supporting different biomolecular modeling tasks. 12. The AtomWorks framework and RF3 model are released with curated training data, code, and model weights, making them accessible for further research and development in the field of biomolecular modeling. 💻Code: github.com/RosettaCommons/at… 📜Paper: biorxiv.org/content/10.1101/… #BiomolecularModeling #AtomWorks #RF3 #StructurePrediction #ProteinDesign #OpenSource #MachineLearning #DeepLearning #ComputationalBiology

De novo design of miniprotein agonists and antagonists targeting G protein-coupled receptors 🚀 New preprint from David Baker!🚀 1. This paper introduces a computational and experimental approach for designing miniproteins targeting G protein-coupled receptors (GPCRs) with high affinity, potency, and selectivity. The method combines RFdiffusion, MetaGen, and Receptor Diversion (RD) microscopy-based high-throughput screening. 2. Two design strategies are employed: (1) Motif-directed RFdiffusion to target specific receptor hotspots for deep pocket interaction. (2) MetaGen, utilizing diverse structural scaffolds generated from AlphaFold-derived metaproteomes. 3. The Receptor Diversion (RD) screening platform operates directly in human cells, eliminating the need for receptor purification and ensuring compatibility with complex cellular environments. This technique enables efficient screening of up to 100,000 designs. 4. Designed miniprotein agonists successfully activated MRGPRX1, a GPCR associated with itch and pain, with two binders demonstrating full and partial agonistic activity, supported by cryo-EM structures confirming their binding accuracy. 5. The study also developed antagonists targeting CXCR4, GLP1R, GIPR, GCGR, and CGRPR, with nanomolar to picomolar affinities. Notably, the CGRPR antagonist dC2_049 displayed exceptional selectivity and potency. 6. Cryo-EM structures confirmed the designed antagonists’ ability to sterically occlude binding sites and prevent receptor activation, offering new opportunities for GPCR-targeted therapeutics. 7. The successful combination of computational design and high-throughput screening demonstrates a promising route for generating novel protein-based therapeutics targeting GPCRs. 📜Paper: biorxiv.org/content/10.1101/… #GPCR #ProteinDesign #DrugDiscovery #RFdiffusion #MetaGen #ReceptorDiversion #Bioinformatics #MachineLearning #AI

DisDock: A Deep Learning Method for Metal Ion-Protein Redocking 1. DisDock introduces a deep learning model designed to predict protein-metal docking with high accuracy, particularly focusing on the interaction of metal ions with proteins using a distance-based approach. 2. Unlike traditional methods, DisDock leverages a U-net architecture combined with self-attention modules, enhancing model performance by using geometric information to better understand atom interactions in metal-protein complexes. 3. The model predicts protein-metal docking by learning the pairwise distances between atoms in the protein and metal ion, allowing it to generate realistic metal binding locations based on protein structures. 4. DisDock’s ability to predict metal binding sites from rigid protein structures offers a powerful tool for studying metalloproteins, especially given its ability to work without prior structural data. 5. Training on a high-quality metalloprotein dataset from the Mother of All Databases (MOAD), DisDock outperforms existing methods such as AutoDock Vina and BioMetAll in terms of docking accuracy, achieving better Euclidean distance and higher success rates. 6. One of DisDock’s key innovations is the integration of an axial-attention module to handle the challenges of self-attention in high-resolution protein images, boosting computational efficiency and predictive accuracy. 7. The model demonstrates flexibility by allowing predictions of metal ion locations even with minimal prior knowledge, showcasing its potential for broader applications in protein structure and function studies. 8. Through extensive testing on a variety of metal ions, DisDock shows that it can predict metal ion locations with high precision, making it an essential tool for understanding the molecular interactions in metalloproteins. 📜Paper: doi.org/10.1002/prot.26791 #DeepLearning #MetalIonDocking #Proteomics #ComputationalBiology #Metalloproteins #Bioinformatics #MachineLearning