RNA-protein binding motifs mining with a new hybrid deep learning based cross-domain knowledge integration approach

By Xiaoyong Pan, Hongbin Shen. 2016

177 citations recorded by the source corpus.

RNAs play key roles in cells through the interactions with proteins known as the RNA-binding proteins (RBP) and their binding motifs enable crucial understanding of the post-transcriptional regulation of RNAs. How the RBPs correctly recognize the target RNAs and why they bind specific positions is still far from clear. Machine learning-based algorithms are widely acknowledged to be capable of speeding up this process. Although many automatic tools have been developed to predict the RNA-protein binding sites from the rapidly growing multi-resource data, e.g. sequence, structure, their domain specific features and formats have posed significant computational challenges. One of current difficulties is that the cross-source shared common knowledge is at a higher abstraction level beyond the observed data, resulting in a low efficiency of direct integration of observed data across domains. The other difficulty is how to interpret the prediction results. Existing approaches tend to terminate after outputting the potential discrete binding sites on the sequences, but how to assemble them into the meaningful binding motifs is a topic worth of further investigation. In viewing of these challenges, we propose a deep learning-based framework (iDeep) by using a novel hybrid convolutional neural network and deep belief network to predict the RBP interaction sites and motifs on RNAs. This new protocol is featured by transforming the original observed data into a high-level abstraction feature space using multiple layers of learning blocks, where the shared representations across different domains are integrated. To validate our iDeep method, we performed experiments on 31 large-scale CLIP-seq datasets, and our results show that by integrating multiple sources of data, the average AUC can be improved by 8% compared to the best single-source-based predictor; and through cross-domain knowledge integration at an abstraction level, it outperforms the state-of-the-art predictors by 6%. Besides the overall enhanced prediction performance, the convolutional neural network module embedded in iDeep is also able to automatically capture the interpretable binding motifs for RBPs. Large-scale experiments demonstrate that these mined binding motifs agree well with the experimentally verified results, suggesting iDeep is a promising approach in the real-world applications. The iDeep framework not only can achieve promising performance than the state-of-the-art predictors, but also easily capture interpretable binding motifs. iDeep is available at http://www.csbio.sjtu.edu.cn/bioinf/iDeep

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

Applications of Deep Learning and Reinforcement Learning to Biological Data (2017, methodology): “sing CLIP AUC values, DBCNN outperforms Ray et al. (0.825 vs. 0.759) [93], multimodal DBN outperforms GraphProt (0.902 vs. 0.887) [84], and DBN-CNN hybrid outperforms NMF based methods (0.9 vs. 0.85) [82]. Also, in predicting miRNA precursor (Fig. 2G), RNN with DA outperformed other non-DL methods (”
The lncLocator: a subcellular localization predictor for long non‐coding RNAs based on a stacked ensemble classifier (2018, unknown): “…structure prediction (Heffernan et al., 2015; Spencer et al., 2015), contact map prediction (Di Lena et al., 2012), lncRNA recognition (Fan and Zhang, 2015), RNA-protein binding motifs identification (Pan and Shen, 2017), ncRNA-protein interaction sequential pattern mining (Pan et al., 2016), etc.”
Deep Learning in Mining Biological Data (2020, methodology): “n et al. to predict responses of human cells using model organisms [57]. Pan et al. applied a hybrid CNN-DBN model on RNAs for the prediction of RNA binding protein (RBP) interaction sites and motifs [58], and Alipanahi et al. used CNN to predict sequence specificities of [D/R]BPs [65]. Denas and Ta”
Deep Learning in Pharmacogenomics: From Gene Regulation to Patient Stratification (2018, background): “ueller/deepcpg Identifies known and de novo sequence motifs that are predictive for DNA methylation levels or methylation variability, and to estimate the effect of single-nucleotide mutations. iDeep [69] and iDeepS [70] http://www.csbio.sjtu.edu. cn/bioinf/iDeep/ https://github.com/xypan1 232/iDeep”