LPI-BLS: Predicting lncRNA–protein interactions with a broad learning system-based stacked ensemble classifier

Many experiment results show that long non-coding RNAs (lncRNAs) play crucial roles in many biological processes, implementing their functions through interaction with RNA-binding proteins (RBPs). Considering that the experimental identification of lncRNA–protein interactions is expensive and time-consuming, many computational methods are proposed to uncover the potential lncRNA–protein interactions. In this study, we develop a novel computational method (namely LPI-BLS) to predict the lncRNA–protein interactions by using the broad learning system and building a stacked ensemble classifier with a logistical regression model. LPI-BLS first adopts the broad learning system to predict the lncRNA–protein interactions. Broad learning system is an alternative way of learning in deep structure and a flat network with few parameters. Then, the results of multiple individual broad learning systems are fed into the stacked ensemble classifier built with a logistical regression to further improve the predictive performance. Compared with other state-of-the-art methods in 5-fold cross-validation test, LPI-BLS has the best performance with the accuracy of 0.902 on RPI488 dataset, the average accuracy of 0.927 on RPI7317 dataset. The results in the independent test also show that our LPI-BLS can effectively predict the lncRNA–protein interactions. The source code can be freely downloaded from https://github.com/NWPU-903PR/LPI_BLS .