1
|
Ouyang D, Liang Y, Wang J, Li L, Ai N, Feng J, Lu S, Liao S, Liu X, Xie S. HGCLAMIR: Hypergraph contrastive learning with attention mechanism and integrated multi-view representation for predicting miRNA-disease associations. PLoS Comput Biol 2024; 20:e1011927. [PMID: 38652712 PMCID: PMC11037542 DOI: 10.1371/journal.pcbi.1011927] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/20/2023] [Accepted: 02/19/2024] [Indexed: 04/25/2024] Open
Abstract
Existing studies have shown that the abnormal expression of microRNAs (miRNAs) usually leads to the occurrence and development of human diseases. Identifying disease-related miRNAs contributes to studying the pathogenesis of diseases at the molecular level. As traditional biological experiments are time-consuming and expensive, computational methods have been used as an effective complement to infer the potential associations between miRNAs and diseases. However, most of the existing computational methods still face three main challenges: (i) learning of high-order relations; (ii) insufficient representation learning ability; (iii) importance learning and integration of multi-view embedding representation. To this end, we developed a HyperGraph Contrastive Learning with view-aware Attention Mechanism and Integrated multi-view Representation (HGCLAMIR) model to discover potential miRNA-disease associations. First, hypergraph convolutional network (HGCN) was utilized to capture high-order complex relations from hypergraphs related to miRNAs and diseases. Then, we combined HGCN with contrastive learning to improve and enhance the embedded representation learning ability of HGCN. Moreover, we introduced view-aware attention mechanism to adaptively weight the embedded representations of different views, thereby obtaining the importance of multi-view latent representations. Next, we innovatively proposed integrated representation learning to integrate the embedded representation information of multiple views for obtaining more reasonable embedding information. Finally, the integrated representation information was fed into a neural network-based matrix completion method to perform miRNA-disease association prediction. Experimental results on the cross-validation set and independent test set indicated that HGCLAMIR can achieve better prediction performance than other baseline models. Furthermore, the results of case studies and enrichment analysis further demonstrated the accuracy of HGCLAMIR and unconfirmed potential associations had biological significance.
Collapse
Affiliation(s)
- Dong Ouyang
- Peng Cheng Laboratory, Shenzhen, China
- School of Biomedical Engineering, Guangdong Medical University, Dongguan, China
| | - Yong Liang
- Peng Cheng Laboratory, Shenzhen, China
- Pazhou Laboratory (Huangpu), Guangzhou, China
| | - Jinfeng Wang
- College of Mathematics and Informatics, South China Agricultural University, Guangzhou, China
| | - Le Li
- School of Computer Science and Engineering, Faculty of Innovation Engineering, Macau University of Science and Technology, Macau, China
| | - Ning Ai
- School of Computer Science and Engineering, Faculty of Innovation Engineering, Macau University of Science and Technology, Macau, China
| | - Junning Feng
- School of Computer Science and Engineering, Faculty of Innovation Engineering, Macau University of Science and Technology, Macau, China
| | - Shanghui Lu
- School of Computer Science and Engineering, Faculty of Innovation Engineering, Macau University of Science and Technology, Macau, China
| | - Shuilin Liao
- School of Computer Science and Engineering, Faculty of Innovation Engineering, Macau University of Science and Technology, Macau, China
| | - Xiaoying Liu
- Computer Engineering Technical College, Guangdong Polytechnic of Science and Technology, Zhuhai, China
| | - Shengli Xie
- Guangdong-HongKong-Macao Joint Laboratory for Smart Discrete Manufacturing, Guangzhou, China
| |
Collapse
|
2
|
Chen K, Zhou Y, Ding M, Wang Y, Ren Z, Yang Y. Self-supervised learning on millions of primary RNA sequences from 72 vertebrates improves sequence-based RNA splicing prediction. Brief Bioinform 2024; 25:bbae163. [PMID: 38605640 PMCID: PMC11009468 DOI: 10.1093/bib/bbae163] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/04/2024] [Revised: 02/22/2024] [Accepted: 03/19/2024] [Indexed: 04/13/2024] Open
Abstract
Language models pretrained by self-supervised learning (SSL) have been widely utilized to study protein sequences, while few models were developed for genomic sequences and were limited to single species. Due to the lack of genomes from different species, these models cannot effectively leverage evolutionary information. In this study, we have developed SpliceBERT, a language model pretrained on primary ribonucleic acids (RNA) sequences from 72 vertebrates by masked language modeling, and applied it to sequence-based modeling of RNA splicing. Pretraining SpliceBERT on diverse species enables effective identification of evolutionarily conserved elements. Meanwhile, the learned hidden states and attention weights can characterize the biological properties of splice sites. As a result, SpliceBERT was shown effective on several downstream tasks: zero-shot prediction of variant effects on splicing, prediction of branchpoints in humans, and cross-species prediction of splice sites. Our study highlighted the importance of pretraining genomic language models on a diverse range of species and suggested that SSL is a promising approach to enhance our understanding of the regulatory logic underlying genomic sequences.
Collapse
Affiliation(s)
- Ken Chen
- School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, China
| | - Yue Zhou
- Peng Cheng Laboratory, Shenzhen, China
| | - Maolin Ding
- School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, China
| | - Yu Wang
- Peng Cheng Laboratory, Shenzhen, China
| | | | - Yuedong Yang
- School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, China
- Key Laboratory of Machine Intelligence and Advanced Computing (Sun Yat-sen University), Ministry of Education, China
| |
Collapse
|
3
|
Shao RW, Wu JW, Wang ZX, Xu H, Yang HQ, Cheng Q, Cui TJ. Macroscopic model and statistical model to characterize electromagnetic information of a digital coding metasurface. Natl Sci Rev 2024; 11:nwad299. [PMID: 38312383 PMCID: PMC10833471 DOI: 10.1093/nsr/nwad299] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/09/2023] [Revised: 11/05/2023] [Accepted: 11/22/2023] [Indexed: 02/06/2024] Open
Abstract
A digital coding metasurface is a platform connecting the digital space and electromagnetic wave space, and has therefore gained much attention due to its intriguing value in reshaping wireless channels and realizing new communication architectures. Correspondingly, there is an urgent need for electromagnetic information theory that reveals the upper limit of communication capacity and supports the accurate design of metasurface-based communication systems. To this end, we propose a macroscopic model and a statistical model of the digital coding metasurface. The macroscopic model uniformly accommodates both digital and electromagnetic aspects of the meta-atoms and predicts all possible scattered fields of the digital coding metasurface based on a small number of simulations or measurements. Full-wave simulations and experimental results show that the macroscopic model is feasible and accurate. A statistical model is further proposed to correlate the mutual coupling between meta-atoms with covariance and to calculate the entropy of the equivalent currents of digital coding metasurface. These two models can help reconfigurable intelligent surfaces achieve more accurate beamforming and channel estimation, and thus improve signal power and coverage. Moreover, the models will encourage the creation of a precoding codebook in metasurface-based direct digital modulation systems, with the aim of approaching the upper limit of channel capacity. With these two models, the concepts of current space and current entropy, as well as the analysis of information loss from the coding space to wave space, is established for the first time, helping to bridge the gap between the digital world and the physical world, and advancing developments of electromagnetic information theory and new-architecture wireless systems.
Collapse
Affiliation(s)
- Rui Wen Shao
- Institute of Electromagnetic Space, Southeast University, Nanjing 210096, China
- State Key Laboratory of Millimeter Waves, Southeast University, Nanjing 210096, China
| | - Jun Wei Wu
- Institute of Electromagnetic Space, Southeast University, Nanjing 210096, China
- State Key Laboratory of Millimeter Waves, Southeast University, Nanjing 210096, China
- Peng Cheng Laboratory, Shenzhen 518055, China
- Pazhou Laboratory (Huangpu), Guangzhou 510555, China
| | - Zheng Xing Wang
- Institute of Electromagnetic Space, Southeast University, Nanjing 210096, China
- State Key Laboratory of Millimeter Waves, Southeast University, Nanjing 210096, China
| | - Hui Xu
- Institute of Electromagnetic Space, Southeast University, Nanjing 210096, China
- State Key Laboratory of Millimeter Waves, Southeast University, Nanjing 210096, China
| | - Han Qing Yang
- Institute of Electromagnetic Space, Southeast University, Nanjing 210096, China
- State Key Laboratory of Millimeter Waves, Southeast University, Nanjing 210096, China
| | - Qiang Cheng
- Institute of Electromagnetic Space, Southeast University, Nanjing 210096, China
- State Key Laboratory of Millimeter Waves, Southeast University, Nanjing 210096, China
| | - Tie Jun Cui
- Institute of Electromagnetic Space, Southeast University, Nanjing 210096, China
- State Key Laboratory of Millimeter Waves, Southeast University, Nanjing 210096, China
- Peng Cheng Laboratory, Shenzhen 518055, China
- Pazhou Laboratory (Huangpu), Guangzhou 510555, China
| |
Collapse
|
4
|
Ren Y, Li P, Liu Z, Chen Z, Chen YL, Peng C, Liu J. Low-threshold nanolasers based on miniaturized bound states in the continuum. Sci Adv 2022; 8:eade8817. [PMID: 36563161 PMCID: PMC9788758 DOI: 10.1126/sciadv.ade8817] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Grants] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 09/14/2022] [Accepted: 11/29/2022] [Indexed: 06/17/2023]
Abstract
The pursuit of compact lasers with low thresholds has imposed strict requirements on tight light confinements with minimized radiation losses. Bound states in the continuum (BICs) have been recently demonstrated as an effective mechanism to trap light. However, most reported BIC lasers are still bulky due to the absence of in-plane light confinement. Here, we combine BICs and photonic bandgaps to realize three-dimensional light confinements, as referred to miniaturized BICs (mini-BICs). We demonstrate highly compact active mini-BIC resonators with a record high-quality (Q) factor of up to 32,500, which enables single-mode lasing with the lowest threshold of 80 W/cm2 among the reported BIC lasers. In addition, photon statistics measurements further confirm the occurrence of the stimulated emission in our devices. Our work reveals a path toward compact BIC lasers with ultralow power consumption and potentially boosts the applications in cavity quantum electrodynamics, nonlinear optics, and integrated photonics.
Collapse
Affiliation(s)
- Yuhao Ren
- State Key Laboratory of Optoelectronic Materials and Technologies, School of Physics, Sun Yat-sen University, Guangzhou 510275, China
| | - Peishen Li
- State Key Laboratory of Advanced Optical Communication System and Networks, School of Electronics and Frontiers Science Center for Nano-optoelectronics, Peking University, Beijing 100871, China
| | - Zhuojun Liu
- State Key Laboratory for Mesoscopic Physics and Frontiers Science Center for Nano-optoelectronics, School of Physics, Peking University, Beijing 100871, China
| | - Zihao Chen
- State Key Laboratory of Advanced Optical Communication System and Networks, School of Electronics and Frontiers Science Center for Nano-optoelectronics, Peking University, Beijing 100871, China
| | - You-Ling Chen
- State Key Laboratory of Integrated Optoelectronics, Institute of Semiconductors, Chinese Academy of Sciences, Beijing 100083, China
| | - Chao Peng
- State Key Laboratory of Advanced Optical Communication System and Networks, School of Electronics and Frontiers Science Center for Nano-optoelectronics, Peking University, Beijing 100871, China
- Peng Cheng Laboratory, Shenzhen 518055, China
| | - Jin Liu
- State Key Laboratory of Optoelectronic Materials and Technologies, School of Physics, Sun Yat-sen University, Guangzhou 510275, China
| |
Collapse
|