Predicting CircRNA-Disease Associations via Feature Convolution Learning With Heterogeneous Graph Attention Network

IEEE Journal of Biomedical and Health Informatics, Journal Year: 2023, Volume and Issue: 27(6), P. 3072 - 3082

Published: March 23, 2023

Exploring the relationship between circular RNA (circRNA) and disease is beneficial for revealing mechanisms of pathogenesis. However, a blind search all possible associations circRNAs diseases through biological experiments time-consuming. Although some prediction methods have been proposed, they still limitations. In this study, novel computational framework, called GATCL2CD, proposed to forecast unknown circRNA-disease (CDAs). First, we calculate Gaussian interactive profile kernel (GIP) similarity semantic diseases, circRNA sequence function similarity, GIPs circRNAs. Then, combine them construct heterogeneous graph. Thereafter, GATCL2CD proposes feature convolution learning that uses multi-head dynamic attention mechanism obtain different aggregated representations features correspond nodes in it extracts rich higher-order from stacked each node by using single-layer convolutional neural network with filter kernels sizes. Finally, pairwise element-wise product operation implemented capture interactions representations, multilayer perceptron introduced as an efficient classifier inferring potential CDAs. Major experimental results under 5-fold cross-validation (5-fold CV) on three datasets show superior five other state-of-the-art methods. Furthermore, case studies demonstrate suitability useful tool identifying disease-related

Language: Английский

---

Deep learning models for disease-associated circRNA prediction: a review

Briefings in Bioinformatics, Journal Year: 2022, Volume and Issue: 23(6)

Published: Aug. 24, 2022

Abstract Emerging evidence indicates that circular RNAs (circRNAs) can provide new insights and potential therapeutic targets for disease diagnosis treatment. However, traditional biological experiments are expensive time-consuming. Recently, deep learning with a more powerful ability representation enables it to be promising technology predicting disease-associated circRNAs. In this review, we mainly introduce the most popular databases related circRNA, summarize three types of learning-based circRNA-disease associations prediction methods: feature-generation-based, type-discrimination hybrid-based methods. We further evaluate seven representative models on benchmark ground truth both balance imbalance classification tasks. addition, discuss advantages limitations each type method highlight suggested applications future research.

Language: Английский

---

A computational model of circRNA-associated diseases based on a graph neural network: prediction and case studies for follow-up experimental validation

BMC Biology, Journal Year: 2024, Volume and Issue: 22(1)

Published: Jan. 29, 2024

Abstract Background Circular RNAs (circRNAs) have been confirmed to play a vital role in the occurrence and development of diseases. Exploring relationship between circRNAs diseases is far-reaching significance for studying etiopathogenesis treating To this end, based on graph Markov neural network algorithm (GMNN) constructed our previous work GMNN2CD, we further considered multisource biological data that affects association circRNA disease developed an updated web server CircDA human hepatocellular carcinoma (HCC) tissue verify prediction results CircDA. Results built Tumarkov-based deep learning framework. The regards biomolecules as nodes interactions molecules edges, reasonably abstracts multiomics data, models them heterogeneous biomolecular network, which can reflect complex different biomolecules. Case studies using literature from HCC, cervical, gastric cancers demonstrate predictor identify missing associations known diseases, quantitative real-time PCR (RT-qPCR) experiment HCC samples, it was found five were significantly differentially expressed, proved predict related new circRNAs. Conclusions This efficient computational case analysis with sufficient feedback allows us circRNA-associated disease-associated Our provides method provide guidance certain For ease use, online ( http://server.malab.cn/CircDA ) provided, code open-sourced https://github.com/nmt315320/CircDA.git convenience improvement.

Language: Английский

---

metaCDA: A Novel Framework for CircRNA-Driven Drug Discovery Utilizing Adaptive Aggregation and Meta-Knowledge Learning

Journal of Chemical Information and Modeling, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 12, 2025

In the emerging field of RNA drugs, circular (circRNA) has attracted much attention as a novel multifunctional therapeutic target. Delving deeper into intricate interactions between circRNA and disease is critical for driving drug discovery efforts centered around circRNAs. Current computational methods face two significant limitations: lack aggregate information in heterogeneous graph networks higher-order fusion information. To this end, we present approach, metaCDA, which utilizes meta-knowledge adaptive learning to improve accuracy association predictions addresses limitations both. We calculate multiple similarity measures circRNA, construct based on these, apply meta-networks extract from graph, so that constructed maps have contrast enhancement Then, nodal aggregation system, integrates multihead mechanism mechanism, achieve accurate capture conducted extensive experiments, results show metaCDA outperforms existing state-of-the-art models can effectively predict disease-associated opening up new prospects circRNA-driven discovery.

Language: Английский

---

Predicting CircRNA-Disease Associations Through Linear Neighborhood Label Propagation Method

IEEE Access, Journal Year: 2019, Volume and Issue: 7, P. 83474 - 83483

Published: Jan. 1, 2019

Identification of circRNA-disease associations provides insight into the mechanism that circRNAs cause diseases. Wet experimental identification is time-consuming and labor-intensive, thus developing computational methods for association prediction an urgent task. In this paper, we propose a linear neighborhood label propagation method to predict associations, named CD-LNLP. First, CD-LNLP uses profiles based on known calculate circRNA-circRNA similarities disease-disease similarities. Next, implements similarity-based graph respectively associations. Finally, combine outputs from model produce results. experiments, achieves impressive performance with AUPR score 0.4487 AUC 0.9007 outperforms outstanding baseline (collaborative filter method, KATZ, nonnegative matrix factorization resource allocation method) state-of-the-art MRLDC. The case studies show identifies novel which are validated by up-to-date databases literature respectively. conclusion, promising predicting

Language: Английский

---

Integrating random walk with restart and k-Nearest Neighbor to identify novel circRNA-disease association

Scientific Reports, Journal Year: 2020, Volume and Issue: 10(1)

Published: Feb. 6, 2020

Abstract CircRNA is a special type of non-coding RNA, which closely related to the occurrence and development many complex human diseases. However, it time-consuming expensive determine circRNA-disease associations through experimental methods. Therefore, based on existing databases, we propose method named RWRKNN, integrates random walk with restart (RWR) k-nearest neighbors (KNN) predict between circRNAs Specifically, apply RWR algorithm weighting features global network topology information, employ KNN classify features. Finally, prediction scores each pair are obtained. As demonstrated by leave-one-out, 5-fold cross-validation 10-fold cross-validation, RWRKNN achieves AUC values 0.9297, 0.9333 0.9261, respectively. And case studies show that predicted can be successfully demonstrated. In conclusion, useful for predicting associations.

Language: Английский

---

MGRCDA: Metagraph Recommendation Method for Predicting CircRNA–Disease Association

IEEE Transactions on Cybernetics, Journal Year: 2021, Volume and Issue: 53(1), P. 67 - 75

Published: July 8, 2021

Clinical evidence began to accumulate, suggesting that circRNAs can be novel therapeutic targets for various diseases and play a critical role in human health. However, limited by the complex mechanism of circRNA, it is difficult quickly large-scale explore relationship between disease circRNA wet-lab experiment. In this work, we design new computational model MGRCDA on account metagraph recommendation theory predict potential circRNA-disease associations. Specifically, first regard association prediction problem as system problem, series metagraphs according heterogeneous biological networks; then extract semantic information Gaussian interaction profile kernel (GIPK) similarity network attributes; finally, iterative search algorithm used calculate scores pair. On gold standard dataset circR2Disease, achieved accuracy 92.49% with an area under ROC curve 0.9298, which significantly higher than other state-of-the-art models. Furthermore, among top 30 disease-related recommended model, 25 have been verified latest published literature. The experimental results prove feasible efficient, recommend reliable candidates further experiment reduce scope

Language: Английский

---

A comprehensive survey on computational methods of non-coding RNA and disease association prediction

Briefings in Bioinformatics, Journal Year: 2020, Volume and Issue: 22(4)

Published: Nov. 2, 2020

The studies on relationships between non-coding RNAs and diseases are widely carried out in recent years. A large number of experimental methods technologies producing biological data have also been developed. However, due to their high labor cost production time, nowadays, calculation-based methods, especially machine learning deep received a lot attention used commonly solve these problems. From computational point view, this survey mainly introduces three common RNAs, i.e. miRNAs, lncRNAs circRNAs, the related for predicting association with diseases. First, mainstream databases above introduced detail. Then, we present several RNA similarity disease calculations. Later, investigate ncRNA-disease prediction details classify into five types: network propagating, recommend system, matrix completion, learning. Furthermore, provide summary applications types associations respectively. Finally, advantages limitations various identified, future researches challenges discussed.

Language: Английский

---

SGANRDA: semi-supervised generative adversarial networks for predicting circRNA–disease associations

Briefings in Bioinformatics, Journal Year: 2021, Volume and Issue: 22(5)

Published: March 16, 2021

Emerging research shows that circular RNA (circRNA) plays a crucial role in the diagnosis, occurrence and prognosis of complex human diseases. Compared with traditional biological experiments, computational method fusing multi-source data to identify association between circRNA disease can effectively reduce cost save time. Considering limitations existing models, we propose semi-supervised generative adversarial network (GAN) model SGANRDA for predicting circRNA-disease association. This first fused natural language features sequence semantics, Gaussian interaction profile kernel, then used all pairs pre-train GAN network, fine-tune parameters through labeled samples. Finally, extreme learning machine classifier is employed obtain prediction result. previous supervision model, innovatively introduced sequences utilized information during pre-training process. step increase content feature some extent impact too few known associations on performance. obtained AUC scores 0.9411 0.9223 leave-one-out cross-validation 5-fold cross-validation, respectively. Prediction results benchmark dataset show outperforms other models. In addition, 25 top 30 highest case studies were verified by recent literature. These experimental demonstrate useful predict provide reliable candidates experiments.

Language: Английский

---

A machine learning framework based on multi-source feature fusion for circRNA-disease association prediction

Briefings in Bioinformatics, Journal Year: 2022, Volume and Issue: 23(5)

Published: Sept. 1, 2022

Abstract Circular RNAs (circRNAs) are involved in the regulatory mechanisms of multiple complex diseases, and identification their associations is critical to diagnosis treatment diseases. In recent years, many computational methods have been designed predict circRNA-disease associations. However, most existing rely on single correlation data. Here, we propose a machine learning framework for association prediction, called MLCDA, which effectively fuses sources heterogeneous information including circRNA sequences disease ontology. Comprehensive evaluation gold standard dataset showed that MLCDA can successfully capture relationships between circRNAs diseases accurately potential addition, results case studies real data show significantly outperforms other methods. serve as useful tool providing mechanistic insights research thus facilitating progress treatment.

Language: Английский

---