4.7 Article

Computational recognition for long non-coding RNA (lncRNA): Software and databases

期刊

BRIEFINGS IN BIOINFORMATICS
卷 18, 期 1, 页码 9-27

出版社

OXFORD UNIV PRESS
DOI: 10.1093/bib/bbv114

关键词

long non-coding RNA; bioinformatics; algorithms; databases

资金

  1. JSPS
  2. JST CREST
  3. PRESTO, JST
  4. JSPS KAKENHI [24300054, 26-381]
  5. Grants-in-Aid for Scientific Research [14J00381, 16H02868] Funding Source: KAKEN

向作者/读者索取更多资源

Since the completion of the Human Genome Project, it has been widely established that most DNA is not transcribed into proteins. These non-protein-coding regions are believed to be moderators within transcriptional and post-transcriptional processes, which play key roles in the onset of diseases. Long non-coding RNAs (lncRNAs) are generally lacking in conserved motifs typically used for detection and thus hard to identify, but nonetheless present certain characteristic features that can be exploited by bioinformatics methods. By combining lncRNA detection with known miRNA, RNA-binding protein and chromatin interaction, current tools are able to recognize and functionally annotate large number of lncRNAs. This review discusses databases and platforms dedicated to cataloging and annotating lncRNAs, as well as tools geared at discovering novel sequences. We emphasize the issues posed by the diversity of lncRNAs and their complex interaction mechanisms, as well as technical issues such as lack of unified nomenclature. We hope that this wide overview of existing platforms and databases might help guide biologists toward the tools they need to analyze their experimental data, while our discussion of limitations and of current lncRNA-related methods may assist in the development of new computational tools.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.7
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

Article Biochemical Research Methods

BERTMeSH: deep contextual representation learning for large-scale high-performance MeSH indexing with full text

Ronghui You, Yuxuan Liu, Hiroshi Mamitsuka, Shanfeng Zhu

Summary: With the rapid increase in biomedical articles, the need for large-scale automatic Medical Subject Headings (MeSH) indexing has grown significantly. This study proposes a computationally lighter method, BERTMeSH, for MeSH indexing using full text and deep learning, outperforming existing methods like FullMeSH in terms of efficiency and flexibility. BERTMeSH utilizes the state-of-the-art BERT model and a transfer learning strategy to achieve superior performance in indexing.

BIOINFORMATICS (2021)

Article Computer Science, Artificial Intelligence

Reshaped tensor nuclear norms for higher order tensor completion

Kishan Wimalawarne, Hiroshi Mamitsuka

Summary: We investigate optimal conditions for inducing low-rankness of higher order tensors using convex tensor norms with reshaped tensors. Proposed reshaped tensor nuclear norm and reshaped latent tensor nuclear norm for regularization and combining multiple tensors, respectively. Through generalization bounds and experiments, the novel reshaping norms are shown to lead to lower complexities, favorably compared to existing tensor norms.

MACHINE LEARNING (2021)

Article Multidisciplinary Sciences

Eukaryotic virus composition can predict the efficiency of carbon export in the global ocean

Hiroto Kaneko, Romain Blanc-Mathieu, Hisashi Endo, Samuel Chaffron, Tom O. Delmont, Morgan Gaia, Nicolas Henry, Rodrigo Hernandez-Velazquez, Canh Hao Nguyen, Hiroshi Mamitsuka, Patrick Forterre, Olivier Jaillon, Colomban de Vargas, Matthew B. Sullivan, Curtis A. Suttle, Lionel Guidi, Hiroyuki Ogata

Summary: There is a significant association between viral community composition and carbon export efficiency on a global scale, with viruses predicted to infect ecologically important hosts playing a crucial role in this process. These findings suggest that viruses likely act in the carbon pump process at a large scale in a manner dependent on their hosts and ecosystem dynamics.

ISCIENCE (2021)

Article Biochemical Research Methods

XGSEA: CROSS-species gene set enrichment analysis via domain adaptation

Menglan Cai, Canh Hao Nguyen, Hiroshi Mamitsuka, Limin Li

Summary: The study introduces a method called CROSS-species gene set enrichment analysis (XGSEA) to predict the enrichment significance of a given gene set in a target species based on gene expression data from a source species, using domain adaptation and regression analysis to improve accuracy. Experimental results show that XGSEA significantly outperforms three baseline methods in most cases.

BRIEFINGS IN BIOINFORMATICS (2021)

Article Biochemical Research Methods

HPOFiller: identifying missing protein-phenotype associations by graph convolutional network

Lizhi Liu, Hiroshi Mamitsuka, Shanfeng Zhu

Summary: Exploring the relationship between human proteins and abnormal phenotypes is crucial for disease prevention, diagnosis and treatment. HPOFiller, a graph convolutional network-based approach, aims to predict missing HPO annotations and outperforms other state-of-the-art methods through stringent evaluations.

BIOINFORMATICS (2021)

Article Biochemical Research Methods

DeepGraphGO: graph neural network for large-scale, multispecies protein function prediction

Ronghui You, Shuwei Yao, Hiroshi Mamitsuka, Shanfeng Zhu

Summary: DeepGraphGO is a multispecies graph neural network-based method aimed at solving the problem of automated function prediction of proteins. By utilizing protein sequence and high-order protein network information, a single model can be trained for all species, providing more training samples for AFP. Experimental results demonstrate that DeepGraphGO significantly outperforms other state-of-the-art methods, including network-based GeneMANIA, deepNF, and clusDCA.

BIOINFORMATICS (2021)

Review Biochemical Research Methods

Machine learning approaches for drug combination therapies

Betul Guvenc Paltun, Samuel Kaski, Hiroshi Mamitsuka

Summary: Drug combination therapy is a promising strategy for treating complex diseases, especially in cancer patients where knowledge is limited. Machine learning methods offer an effective way to improve therapeutic efficacy and overcome drug resistance. Data integration and experimental comparisons play a crucial role in drug combination analysis.

BRIEFINGS IN BIOINFORMATICS (2021)

Article Computer Science, Artificial Intelligence

Learning subtree pattern importance for Weisfeiler-Lehman based graph kernels

Dai Hai Nguyen, Canh Hao Nguyen, Hiroshi Mamitsuka

Summary: This paper proposes a method for learning the weights of subtree patterns within the framework of WWL kernels, demonstrating its effectiveness in graph classification tasks and showcasing its validity through experiments on synthetic and real-world datasets.

MACHINE LEARNING (2021)

Article Biochemical Research Methods

HPODNets: deep graph convolutional networks for predicting human protein-phenotype associations

Lizhi Liu, Hiroshi Mamitsuka, Shanfeng Zhu

Summary: Deciphering the relationship between human genes/proteins and abnormal phenotypes is crucial for disease prevention and treatment, requiring computational predictions. The HPODNets model, with features including multiple network input, semi-supervised learning, and deep graph convolutional network, outperforms other methods in predicting human protein-phenotype associations.

BIOINFORMATICS (2022)

Article Multidisciplinary Sciences

Crop loss identification at field parcel scale using satellite remote sensing and machine learning

Santosh Hiremath, Samantha Wittke, Taru Palosuo, Jere Kaivosoja, Fulu Tao, Maximilian Proll, Eetu Puttonen, Pirjo Peltonen-Sainio, Pekka Marttinen, Hiroshi Mamitsuka

Summary: This study investigates the feasibility of using satellite images and machine learning models to classify agricultural field parcels into those with and without crop loss. Despite the poor quality of data, the random forest model shows promising results in identifying new crop-loss fields based on reference fields of the same year. There is potential for various applications in efficient agricultural monitoring and verifying crop-loss claims.

PLOS ONE (2021)

Article Biochemical Research Methods

DeepMHCII: a novel binding core-aware deep interaction model for accurate MHC-II peptide binding affinity prediction

Ronghui You, Wei Qu, Hiroshi Mamitsuka, Shanfeng Zhu

Summary: This study proposes a deep learning model, DeepMHCII, based on peptide binding cores and introduces a binding interaction convolution layer to better model the biological interactions between peptides and MHC II molecules. Extensive experiments demonstrate that DeepMHCII outperforms existing methods in terms of performance and can effectively predict binding cores.

BIOINFORMATICS (2022)

Article Biochemical Research Methods

SPARSE: a sparse hypergraph neural network for learning multiple types of latent combinations to accurately predict drug-drug interactions

Duc Anh Nguyen, Canh Hao Nguyen, Peter Petschner, Hiroshi Mamitsuka

Summary: Predicting the side effects of drug-drug interactions is an important task in pharmacology. Existing methods use hypergraph neural networks to learn the relationships between drugs and side effects but cannot accurately represent the multiple mechanisms of side effects. In this paper, we propose SPARSE, a method that encodes the DDI hypergraph and drug features to learn multiple combinations of latent features of drugs and side effects. By controlling model sparsity through a sparse prior, SPARSE achieves superior predictive performance and interpretability advantage.

BIOINFORMATICS (2022)

Article Biochemical Research Methods

DIVERSE: Bayesian Data IntegratiVE Learning for Precise Drug ResponSE Prediction

Betul Guvenc Paltun, Samuel Kaski, Hiroshi Mamitsuka

Summary: Detecting predictive biomarkers from multi-omics data is crucial for precision medicine, but choosing reliable data sources is challenging. We propose the DIVERSE framework which integrates diverse data sets to predict drug responses, and it outperforms other methods in empirical experiments.

IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS (2022)

Article Computer Science, Artificial Intelligence

Central-Smoothing Hypergraph Neural Networks for Predicting Drug-Drug Interactions

Duc Anh Nguyen, Canh Hao Nguyen, Hiroshi Mamitsuka

Summary: Predicting drug-drug interactions (DDIs) involves predicting the side effects of drug pairs using drug information and known side effects. This problem can be solved by predicting labels for each drug pair in a DDI graph, where drugs are nodes and interacting drugs with known labels are edges. Graph neural networks (GNNs) are commonly used for this problem, but they may not perform well for infrequent labels and complicated label relationships. To address this, the authors propose CentSmoothie, a hypergraph neural network (HGNN) that learns representations of drugs and labels using a central-smoothing formulation. Experimental results on simulations and real datasets demonstrate the superiority of CentSmoothie.

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS (2023)

Article Computer Science, Artificial Intelligence

Learning on Hypergraphs With Sparsity

Canh Hao Nguyen, Hiroshi Mamitsuka

Summary: Hypergraph is a generalization of graph for representing high-order relations on a set of objects. To address the issue of irrelevant or noisy data, a sparse learning framework is incorporated into learning on hypergraphs. Sparse smooth formulations are proposed to learn smooth functions and induce sparsity on both hyperedges and nodes of hypergraphs.

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2021)

暂无数据