A novel lncRNA\u2013protein interaction prediction method based on deep forest with cascade forest structure

Xiongfei Tian,Liqian Zhou,Ling Shen,Zhenwu Wang,Lihong Peng

doi:10.1038/s41598-021-98277-1

Xiongfei Tian, Liqian Zhou + Show 3 more

Open Access

https://doi.org/10.1038/s41598-021-98277-1

Copy DOI

Abstract

Long noncoding RNAs (lncRNAs) regulate many biological processes by interacting with corresponding RNA-binding proteins. The identification of lncRNA–protein Interactions (LPIs) is significantly important to well characterize the biological functions and mechanisms of lncRNAs. Existing computational methods have been effectively applied to LPI prediction. However, the majority of them were evaluated only on one LPI dataset, thereby resulting in prediction bias. More importantly, part of models did not discover possible LPIs for new lncRNAs (or proteins). In addition, the prediction performance remains limited. To solve with the above problems, in this study, we develop a Deep Forest-based LPI prediction method (LPIDF). First, five LPI datasets are obtained and the corresponding sequence information of lncRNAs and proteins are collected. Second, features of lncRNAs and proteins are constructed based on four-nucleotide composition and BioSeq2vec with encoder-decoder structure, respectively. Finally, a deep forest model with cascade forest structure is developed to find new LPIs. We compare LPIDF with four classical association prediction models based on three fivefold cross validations on lncRNAs, proteins, and LPIs. LPIDF obtains better average AUCs of 0.9012, 0.6937 and 0.9457, and the best average AUPRs of 0.9022, 0.6860, and 0.9382, respectively, for the three CVs, significantly outperforming other methods. The results show that the lncRNA FTX may interact with the protein P35637 and needs further validation.

Highlights

Long noncoding RNAs regulate many biological processes by interacting with corresponding RNA-binding proteins
Network-based lncRNA–protein Interactions (LPIs) prediction methods, for example, random walk with restart-based m odel[17], linear neighborhood propagation a lgorithm[18], bipartite network projection-based recommendation method[19,20,21], HeteSim algorithm[22], firstly computed Long noncoding RNAs (lncRNAs) similarity and protein similarity based on related biological data, and integrated similarity matrix to heterogeneous lncRNA–protein network, designed network propagation algorithms to score for unknown lncRNA–protein pairs
Features of lncRNAs and proteins are selected by four-nucleotide composition and BioSeq2vec based on their sequences, respectively

Summary

Introduction

Long noncoding RNAs (lncRNAs) regulate many biological processes by interacting with corresponding RNA-binding proteins. Machine learning-based LPI identification methods first extracted features of lncRNAs and proteins and designed a novel machine learning model to compute interaction probabilities for lncRNA–protein pairs. The features of lncRNAs and proteins are extracted based on four-nucleotide composition and the BioSeq2vec methods, respectively.

Results

Conclusion