A novel end-to-end method to predict RNA secondary structure profile based on bidirectional LSTM and residual neural network

Linyu Wang,Hao Zhang,Xiaodan Zhong,Shuo Wang,Yuanning Liu

doi:10.1186/s12859-021-04102-x

Abstract

BackgroundStudies have shown that RNA secondary structure, a planar structure formed by paired bases, plays diverse vital roles in fundamental life activities and complex diseases. RNA secondary structure profile can record whether each base is paired with others. Hence, accurate prediction of secondary structure profile can help to deduce the secondary structure and binding site of RNA. RNA secondary structure profile can be obtained through biological experiment and calculation methods. Of them, the biological experiment method involves two ways: chemical reagent and biological crystallization. The chemical reagent method can obtain a large number of prediction data, but its cost is high and always associated with high noise, making it difficult to get results of all bases on RNA due to the limited of sequencing coverage. By contrast, the biological crystallization method can lead to accurate results, yet heavy experimental work and high costs are required. On the other hand, the calculation method is CROSS, which comprises a three-layer fully connected neural network. However, CROSS can not completely learn the features of RNA secondary structure profile since its poor network structure, leading to its low performance.ResultsIn this paper, a novel end-to-end method, named as “RPRes, was proposed to predict RNA secondary structure profile based on Bidirectional LSTM and Residual Neural Network.ConclusionsRPRes utilizes data sets generated by multiple biological experiment methods as the training, validation, and test sets to predict profile, which can compatible with numerous prediction requirements. Compared with the biological experiment method, RPRes has reduced the costs and improved the prediction efficiency. Compared with the state-of-the-art calculation method CROSS, RPRes has significantly improved performance.

Highlights

Studies have shown that RNA secondary structure, a planar structure formed by paired bases, plays diverse vital roles in fundamental life activities and complex diseases
The similarities were that both RPRes and CROSS were based on neural network, and the multiple data sets created by the biological experiment method were adopted for training, validation, and test; in this way, the characters of multiple biological experiment data sets were taken into account
RPRes first extracted the features of all target base data transmitted into the same format output based on a Bidirectional LSTM (Bi-LSTM) layer, and a ResNet was used to classify the output of Bi-LSTM

Summary

Introduction

Studies have shown that RNA secondary structure, a planar structure formed by paired bases, plays diverse vital roles in fundamental life activities and complex diseases. Accurate prediction of secondary structure profile can help to deduce the secondary structure and binding site of RNA. RNA secondary structure profile can be obtained through biological experiment and calculation methods. The chemical reagent method can obtain a large number of prediction data, but its cost is high and always associated with high noise, making it difficult to get results of all bases on RNA due to the limited of sequencing coverage. The accurate prediction of RNA secondary structure profile helps to deduce the RNA secondary structure without length restriction [15] and helps to identify the binding site of RNA, promoting the functional research of RNA

Methods

Results

Discussion

Conclusion