De Novo RNA Tertiary Structure Prediction at Atomic Resolution Using Geometric Potentials from Deep Learning

By Robin Pearce, Gilbert S. Omenn, Yang Zhang. 2022

82 citations recorded by the source corpus.

Experimental characterization of RNA structure remains difficult, especially for non-coding RNAs that are critical to many cellular activities. We developed DeepFoldRNA to predict RNA structures from sequence alone by coupling deep self-attention neural networks with gradient-based folding simulations. The method was tested on two independent benchmark datasets from Rfam families and RNA-Puzzle experiments, where DeepFoldRNA constructed models with an average RMSD=2.69 Å and TM-score=0.743, which outperformed state-of-the-art methods and the best models submitted from the RNA-Puzzles community by a large margin. On average, DeepFoldRNA required ~1 minute to fold medium-sized RNAs, which was ~350-4000 times faster than the leading Monte Carlo simulation approaches. These results demonstrate the major advantage of advanced deep learning techniques to learn more accurate information from evolutionary profiles than knowledge-based potentials derived from simple statistics of the PDB library. The high speed and accuracy of the developed method should enable large-scale atomic-level RNA structure modeling applications.

Subfields: Molecular Biology

Citations in the corpus

• Accurate RNA 3D structure prediction using a language model-based deep learning approach (2022, neutral): “Several top-ranking CASP15 groups and recent published works on RNA 3D structure prediction 17,18,20,21,23 were included in our benchmarking.”
• OpenFold: retraining AlphaFold2 yields new insights into its learning mechanisms and capacity for generalization (2023, unknown): “It is possible that, in earlier phases of training, OpenFold has already recovered aspects of secondary structure not recognized by DSSP on account of imprecise atom positioning.”
• Integrating end-to-end learning with deep geometrical potentials for ab initio RNA structure prediction (2022, unknown): “For this, we combined the end-to-end potentials (from DRfold) and geometry potentials (constructed from DeepFoldRNA predictions) into a new hybrid potential and use it to guide the subsequent structure optimization, while keeping other part of the DRfold procedure unchanged.”