RNA Secondary Structure Prediction By Learning Unrolled Algorithms

By Xinshi Chen, Yu Li, Ramzan Umarov, Xin Gao, Le Song. 2020

137 citations recorded by the source corpus.

In this paper, we propose an end-to-end deep learning model, called E2Efold, for RNA secondary structure prediction which can effectively take into account the inherent constraints in the problem. The key idea of E2Efold is to directly predict the RNA base-pairing matrix, and use an unrolled algorithm for constrained programming as the template for deep architectures to enforce constraints. With comprehensive experiments on benchmark datasets, we demonstrate the superior performance of E2Efold: it predicts significantly better structures compared to previous SOTA (especially for pseudoknotted structures), while being as efficient as the fastest algorithms in terms of inference time.

Subfields: Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Molecular Biology

Citations in the corpus

Accurate RNA 3D structure prediction using a language model-based deep learning approach (2022, neutral): “These applications include predicting protein 3D structures 13 , RNA secondary structures 14,15 and scoring RNA structures generated by other methods 16 .”
RNA secondary structure prediction using deep learning with thermodynamic integration (2020, methodology): “We compared our algorithm, MXfold2, with two recently published DNN-based methods, SPOT-RNA (18) and E2Efold (19).”
Learning to Optimize: A Primer and A Benchmark (2021, methodology): “Therefore, most LSTM-based L2O methods [13; 53; 135; 227; 149; 37; 46; 47] are forced to limit their maximal unroll lengths and to truncate the unrolled optimization (e.”
UFold: fast and accurate RNA secondary structure prediction with deep learning (2020, unknown): “These methods either treat the input as a sequence, utilizing LSTM (39) or transformer encoder (40) to capture long-range interactions between nucleotides (37,41,42).”
Interpretable RNA Foundation Model from Unannotated Data for Highly Accurate RNA Structure and Function Predictions (2022, background): “Nevertheless, the generalization capability of such DL-based methods still remains a problem, as the model architecture is explicitly designed for corresponding tasks and cannot generalize well to unknown RNA types [30].”
Integrating end-to-end learning with deep geometrical potentials for ab initio RNA structure prediction (2022, unknown): “Previous studies (Chen et al., 2020; Singh et al., 2019) have showed the success in learning the RNA secondary structures from”
Recent advances in RNA structurome (2022, background): “machine learning or deep learning to model or predict RNA secondary structures, including E2Efold (Chen et al., 2020), MXfold (Sato et al.”