Tensor decomposition-based unsupervised feature extraction applied to matrix products for multi-view data processing

Taguchi, Y-h.

doi:10.1371/journal.pone.0183933

Cited by 30 publications

(23 citation statements)

References 73 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is Type I Case II tensor (32). Since it is too large to be decomposed, we further transform it into type II tensor as where x jk ∊ ℝ M × K is now not a tensor but a matrix.…”

Section: Methodsmentioning

confidence: 99%

“…In this paper, we propose the application of tensor decomposition (TD) based unsupervised FE (37, 46, 35, 36, 30, 32, 29, 33, 28), which is an advanced method of PCA based unsupervised FE, to scRNA-seq analysis (for more details about PCA based unsupervised FE and TD based unsupervised FE, see the recently published book (38)), especially focusing on the integration of two scRNA-seq profiles, The advantages of TD based unsupervised FE when compared with PCA based unsupervised FE are that the former can integrated more than two gene expression prior to the analysis while the latter can only integrate the results obtained by applying the method to individual data set. In the following, following the previous study (34) where PCA based unsupervised FE was employed, we try to integrate human and mouse midbrain development gene expression profiles to obtain key genes that contribute to this process, by applying TD based unsupervised FE.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Tensor decomposition-Based Unsupervised Feature Extraction Applied to Single-Cell Gene Expression Analysis

Taguchi

Turki

2019

Preprint

Self Cite

View full text Add to dashboard Cite

3Although single cell RNA sequencing (scRNA-seq) technology is newly invented and promising 4 one, because of lack of enough information that labels individual cells, it is hard to interpret 5 the obtained gene expression of each cell. Because of this insufficient information available, 6 unsupervised clustering, e.g., tSNE and UMAP, is usually employed to obtain low dimensional 7 embedding that can help to understand cell-cell relationship. One possible drawback of this 8 strategy is that the outcome is highly dependent upon genes selected for the usage of clustering. 9 In order to fulfill this requirements, there are many methods that performed unsupervised gene 10 selection. In this study, a tensor decomposition (TD) based unsupervised feature extraction 11 (FE) was applied to the integration of two scRNA-seq expression profiles that measure human 12 and mouse midbrain development. TD based unsupervised FE could not only select coincident 13 genes between human and mouse, but also biologically reliable genes. Coincidence between two 14 species as well as biological reliability of selected genes is increased compared with principal 15 component analysis (PCA) based FE applied to the same data set in the previous study. Since 16 PCA based unsupervised FE outperformed other three popular unsupervised gene selection 17 methods, highly variable genes, bimodal genes and dpFeature, TD based unsupervised FE can 18 do so as well. In addition to this, ten transcription factors (TFs) that might regulate selected genes 19 and might contribute to midbrain development are identified. These ten TFs, BHLHE40, EGR1, 20 GABPA, IRF3, PPARG, REST, RFX5, STAT3, TCF7L2, and ZBTB33, were previously reported to 21 be related to brain functions and diseases. TD based unsupervised FE is promising method to 22 integrate two scRNA-seq profiles effectively. 23 Keywords: tensor decomposition, enrichment analysis, sngle cell RNA-sequencing, midbrain development, inter-species analysis 24 INTRODUCTIONSingle cell RNA sequencing (scRNA-seq) (17) is a newly invented technology that enables us to measure 25 amount of RNA in single cell basis. In spite of its promising potential, it is not easy to interpret the 26 measurements. The primary reason of this difficulty is the lack of sufficient information that characterizes 27 1 Y-h. Taguchi TD based FE to single-cell individual cells. In contrast to the huge number of cells measured, which is often as many as several 28 thousands, the number of labeling is limited, e.g., measurement of conditions as well as the amount of 29 expression of key genes measured by fluorescence-activated cell sorting, whose number is typically as little 30 as tens. This prevents us from selecting genes that characterize the individual cell properties. 31In order to deal with samples without suitable number of labelling, unsupervised method is frequently 32 used, since it does not make use of labeling information directly. K-means clustering as well as hierarchical 33 clustering are the popular methodology tha...

show abstract

“…It is Type I Case II tensor (32). Since it is too large to be decomposed, we further transform it into type II tensor as where x jk ∊ ℝ M × K is now not a tensor but a matrix.…”

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Tensor decomposition-Based Unsupervised Feature Extraction Applied to Single-Cell Gene Expression Analysis

Taguchi

Turki

2019

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…We apply TD with higher order singular value decomposition (HOSVD) [3] algorithm to case I type I tensor [4]. Starting from three matrices, 1 mRNA ∈ ℝ 200 ×150 , 2 miRNA ∈ ℝ 184 ×150 , and 3 prot ∈ ℝ 142 ×150 , we generate four mode tensor…”

Section: Tensor Decompositionmentioning

confidence: 99%

Multiomics data analysis using tensor decomposition based unsupervised feature extraction --Comparison with DIABLO--

Taguchi

2019

Preprint

Self Cite

View full text Add to dashboard Cite

Multiomics data analysis is the central issue of genomics science. In spite of that, there are not well defined methods that can integrate multomics data sets, which are formatted as matrices with different sizes. In this paper, I propose the usage of tensor decomposition based unsupervised feature extraction as a data mining tool for multiomics data set. It can successfully integrate miRNA expression, mRNA expression and proteome, which were used as a demonstration example of DIABLO that is the recently proposed advanced method for the integrated analysis of multiomics data set.

show abstract

“…Tensor decomposition is a dimensionality reduction operation on the tensor [20,21]. Similar to principal component analysis and singular value decomposition methods, tensor decomposition methods are devoted to extracting the main structure and compositional information in the original tensor [22,23].…”

Section: Definitions and Notationsmentioning

confidence: 99%

“…The tensor is actually a multidimensional array [24], and we use the Euler script letters ( ) to represent the tensor. We refer to the tensor dimension and number of tensor dimension as modes and order, respectively [22]. Scalar, vector, and matrix are denoted in lowercase ( ), bold lowercase (a), and uppercase letters ( ), correspondingly; the transposition of matrix in ; and unit matrices in .…”

Section: Definitions and Notationsmentioning

confidence: 99%

Learning Users’ Intention of Legal Consultation through Pattern-Oriented Tensor Decomposition with Bi-LSTM

Guo

Zhang

2019

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

Online legal consultation plays an increasingly important role in the modern rule-of-law society. This study aims to understand the intention of legal consultation of users with different language expressions and legal knowledge background. A critical issue is a method through which users’ legal consultation data are classified and the feature of each category is extracted. Traditional classification methods rely considerably on lexical and syntactic features and frequently require strict sentence formatting, which eliminates substantial energy and may not be universally applicable. We aim to extract the patterns of users’ consultation on different categories, which minimally depend on lexical, syntax, and sentence formatting. However, research in this area has rarely been conducted in previous legal advisory service studies. In this study, a classification approach for multiclass users’ intention based on pattern-oriented tensor decomposition and Bi-LSTM is proposed, and each user’s legal consulting statement is expressed as a tensor. Moreover, we propose a pattern-oriented tensor decomposition method that can obtain a core tensor that approximates the patterns of users’ consultation. These patterns can improve the accuracy of classifying users’ intention of legal consultation. We use Bi-LSTM to automatically learn and optimize these patterns. Evidently, Bi-LSTM with a pattern-oriented tensor decomposition layer performs better than a recurrent neural network only. Results show that our method is more accurate than the previous work, and the factor matrix and core tensor calculated by the pattern-oriented tensor decomposition are interpretable.

show abstract

Tensor decomposition-based unsupervised feature extraction applied to matrix products for multi-view data processing

Cited by 30 publications

References 73 publications

Tensor decomposition-Based Unsupervised Feature Extraction Applied to Single-Cell Gene Expression Analysis

Tensor decomposition-Based Unsupervised Feature Extraction Applied to Single-Cell Gene Expression Analysis

Multiomics data analysis using tensor decomposition based unsupervised feature extraction --Comparison with DIABLO--

Learning Users’ Intention of Legal Consultation through Pattern-Oriented Tensor Decomposition with Bi-LSTM

Contact Info

Product

Resources

About