RNA design has been hindered by the limited accuracy of three-dimensional (3D) structure prediction. In this study, we show that intricate RNA structures can be generated with current deep learning tools through accurate de novo design of pseudoknot secondary structures. In an Eterna competition involving 57 pseudoknots, generative artificial intelligence (AI) methods matched experienced human designers in solving most blind challenges, evaluated by single nucleotide-resolution chemical mapping, compensatory mutagenesis, and cryo-electron microscopy. AI-generated molecules with accurate secondary structures formed well-ordered 3D folds stabilized by noncanonical tertiary interactions not modeled during design. Success was guided by an RNet foundation model trained on prior chemical mapping data, suggesting that some difficult RNA design tasks may be tractable without first solving RNA 3D structure prediction.
J. Townley, W. Kladwang, David Baker et al.· Science· 1 citation
Many proteins' biological functions rely on interconversions between multiple conformations occurring at micro- to millisecond (µs-ms) timescales. A lack of standardized, large-scale experimental data has hindered obtaining a more predictive understanding of these motions. After curating >100 Nuclear Magnetic Resonance (NMR) relaxation datasets, we realized an observable for µs-ms dynamics might be hiding in plain sight. Millisecond dynamics can cause NMR signals to broaden beyond detection, leaving some residues not assigned in the chemical shift datasets of ~10,000 proteins deposited in the Biological Magnetic Resonance Data Bank (BMRB) 1. We made the bold assumption that residues missing assignments are exchange-broadened due to µs-ms motions and trained various deep learning models to predict missing assignments. Strikingly, these models also predict exchange measured via NMR relaxation experiments, indicative of µs-ms dynamics. The best of these models, which we named Dyna-1, leverages an intermediate layer of the multimodal language model ESM-32. Notably, dynamics directly linked to biological function, including enzyme catalysis and ligand binding, are particularly well predicted by Dyna-1, which parallels our findings that residues experiencing µs-ms exchange are more conserved. We anticipate the datasets and models presented here will be transformative in unlocking the common language of dynamics and function.
Hannah K. Wayment-Steele, Gina El Nesr, Ramith Hettiarachchi et al.· Nature· 0 citations
Nuclear magnetic resonance (NMR) spectroscopy yields rich residue-level information on biomolecular dynamics and chemical environments, two frontiers for quantitative predictive methods in biochemistry. Decades of data are publicly archived in the Biological Magnetic Resonance Data Bank (BMRB)1, yet in practice, this information remains difficult to access and interpret at scale and within computational workflows. Here we present makeshift, an open-source Python package for accessing, curating, and analyzing NMR datasets. Users can readily retrieve and parse BMRB entries and perform essential analyses such as chemical shift re-referencing, secondary structure propensity prediction, and interpretation of relaxation datasets for dynamics. We re-implemented several widely-used NMR data calculations which were not open-source or available in Python and validated our implementations against the original implementations. By integrating data access, processing, and analysis into a single Python interface, makeshift lowers the barrier for reproducible, scalable analysis and machine learning applications using biomolecular NMR data.
Gina El Nesr, Hannah K. Wayment-Steele· bioRxiv· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
