NPI-RGCNAE: Fast Predicting ncRNA-Protein Interactions Using the Relational Graph Convolutional Network Auto-Encoder

Yu, Han; Shen, Zi-Ang; Du, Pu-Feng

doi:10.1109/jbhi.2021.3122527

Cited by 9 publications

(4 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed IMCN model was trained using the Adam optimizer with 500 epochs and the following settings: (1) The ReLU function is adopted as the non-linear activation of hidden layers. (2) The output dimension of local and global node representations is fixed to the number of classes. The dimensions of hidden layers, learning rate, weight decay, and dropout ratio are searched in {32, 64, 128}, {0.1, 0.05, 0.01}, {0.01, 0.005, 0.001, 0.0005}, and {0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9}, respectively.…”

Section: Experimental Settingsmentioning

confidence: 99%

“…Graphs are useful data structures for capturing relationships between entities in various complex interconnected systems, such as social relationships [1], protein interactions [2], commodity co-purchasing [3], and co-citations [4]. Many fundamental tasks on graphs involve making predictions over nodes, such as predicting labels for unlabeled nodes according to the graph structure and node attributes.…”

Section: Introductionmentioning

confidence: 99%

“…Such outcomes are unfavorable when aiming to learn a discriminative classifier. (2) The class label information of nodes is scarce and concise but plays a significant role in supervising model learning. The relation between nodes and classes should be taken full advantage of for mining valuable supervised signals and optimizing the processes of node embedding and classification.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Individuality-enhanced and multi-granularity consistency-preserving graph neural network for semi-supervised node classification

Liu,

2023

Appl Intell

View full text Add to dashboard Cite

Semi-supervised node classification is an important task that aims at classifying nodes based on the graph structure, node features, and class labels for a subset of nodes. While most graph convolutional networks (GCNs) perform well when an ample number of labeled nodes are available, they often degenerate when the amount of labeled data is limited. To address this problem, we propose a scheme, namely, Individuality-enhanced and Multi-granularity Consistency-preserving graph neural Network (IMCN), which can alleviate the problem of losing individual information within the encoder while providing a reliable supervised signal for learning purposes. First, one simple encoder based on node features only is integrated to enhance node individuality and amend node commonality learned by the GCN-based encoder. Then, three constraints are defined at different levels of granularity, encompassing node embedding agreement, semantic class alignment, and node-to-class distribution identity. They can maintain the consistency between the individuality and commonality of nodes and be leveraged as latent supervised signals for learning representative embeddings. Finally, the trade-off between the individuality and commonality of nodes captured by two encoders is taken into consideration for node classification. Extensive experiments on six real-world datasets have been conducted to validate the superiority of IMCN against state-of-the-art baselines in handling node classification tasks with scarce labeled data. Graphical abstract

show abstract

Section: Experimental Settingsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Individuality-enhanced and multi-granularity consistency-preserving graph neural network for semi-supervised node classification

Liu,

2023

Appl Intell

View full text Add to dashboard Cite

show abstract

“…DM-RPIs extracted sequence characteristics through making full use of stacked auto-encoder networks and trained through multiple base classifier ( Cheng et al, 2019 ). NPI-RGCNAE is proposed by Yu et al utilizing graph convolutional network (GCN) to predict ncRNA-protein interactions, and they developed a novel approach of negative sample selecting ( Yu et al, 2021 ). Although existing computational methods using different RNA and protein features to predict with good performance, these methods may be ineffective due to the features may not available to all RNAs and proteins, particularly facing to new RNA and protein, which have no known interactions with any protein or RNA.…”

Section: Introductionmentioning

confidence: 99%

SAWRPI: A Stacking Ensemble Framework With Adaptive Weight for Predicting ncRNA-Protein Interactions Using Sequence Information

Ren

et al. 2022

Front. Genet.

View full text Add to dashboard Cite

Non-coding RNAs (ncRNAs) take essential effects on biological processes, like gene regulation. One critical way of ncRNA executing biological functions is interactions between ncRNA and RNA binding proteins (RBPs). Identifying proteins, involving ncRNA-protein interactions, can well understand the function ncRNA. Many high-throughput experiment have been applied to recognize the interactions. As a consequence of these approaches are time- and labor-consuming, currently, a great number of computational methods have been developed to improve and advance the ncRNA-protein interactions research. However, these methods may be not available to all RNAs and proteins, particularly processing new RNAs and proteins. Additionally, most of them cannot process well with long sequence. In this work, a computational method SAWRPI is proposed to make prediction of ncRNA-protein through sequence information. More specifically, the raw features of protein and ncRNA are firstly extracted through the k-mer sparse matrix with SVD reduction and learning nucleic acid symbols by natural language processing with local fusion strategy, respectively. Then, to classify easily, Hilbert Transformation is exploited to transform raw feature data to the new feature space. Finally, stacking ensemble strategy is adopted to learn high-level abstraction features automatically and generate final prediction results. To confirm the robustness and stability, three different datasets containing two kinds of interactions are utilized. In comparison with state-of-the-art methods and other results classifying or feature extracting strategies, SAWRPI achieved high performance on three datasets, containing two kinds of lncRNA-protein interactions. Upon our finding, SAWRPI is a trustworthy, robust, yet simple and can be used as a beneficial supplement to the task of predicting ncRNA-protein interactions.

show abstract

Learning from explainable data-driven tunneling graphs: A spatio-temporal graph convolutional network for clogging detection

Gao

Chen

Qin

et al. 2023

Automation in Construction

View full text Add to dashboard Cite

NPI-RGCNAE: Fast Predicting ncRNA-Protein Interactions Using the Relational Graph Convolutional Network Auto-Encoder

Cited by 9 publications

References 53 publications

Individuality-enhanced and multi-granularity consistency-preserving graph neural network for semi-supervised node classification

Individuality-enhanced and multi-granularity consistency-preserving graph neural network for semi-supervised node classification

SAWRPI: A Stacking Ensemble Framework With Adaptive Weight for Predicting ncRNA-Protein Interactions Using Sequence Information

Learning from explainable data-driven tunneling graphs: A spatio-temporal graph convolutional network for clogging detection

Contact Info

Product

Resources

About