Unsupervised Gene Selection Research Articles

Uncertainty measurement for single cell RNA-seq data via Gaussian kernel: Application to unsupervised gene selection

A real-valued information system (RVIS) is an information system (IS) whose information values are real numbers. If the objects, attributes and information values of a RVIS change to cells, genes and gene expression values where gene expression data is single cell RNA-seq (scRNA) data, respectively, then this RVIS is referred to as a single cell gene space (scg-space). Unsupervised gene selection becomes very challenging due to a lack of decision information, which is to select the optimal gene subset that can maintain learning ability without decision information. However, little research has been done on unsupervised gene selection. Uncertainty measurement is a tool of gene selection. In view of this, this paper studies uncertainty measurement in an scg-space via Gaussian kernel and explores its application for unsupervised gene selection. In the first place, the distance between two cells in a given subspace is constructed. In the next place, the fuzzy Tcos-equivalence relation induced by this subspace is obtained employing Gaussian kernel. After that, measures of uncertainty for an scg-space are investigated. Lastly, gene selection algorithms in an scg-space are presented by using the proposed information entropy and information granularity. The presented algorithms are applied to clustering analyses of scRNA data. Multiple publicly available scRNA data sets are employed to evaluate the gene selection performances of the presented algorithms, while two commonly-used clustering methods, kmeans and AGNES, are utilized to obtain four metrics such as Silhouette Coefficient (SC), Davies–Bouldin Index (DBI), Fowlkes and Mallows Index (FMI), Normalized Mutual Information (NMI) . The clustering results demonstrated that the presented algorithms can lower significantly the number genes selected, achieve the better SC, DBI, FMI and NMI. They also show that the presented algorithms are superior to raw data and PCA and NMF regardless of using kmeans or AGNES clustering. This also indirectly demonstrates that the granulation measure and information entropy can effectively evaluate the uncertainty of an scg-space.

An unsupervised gene selection method based on multivariate normalized mutual information of genes

Gene expression data analysis has always been challenging due to complex and high-dimensional samples and genes. Generally, the number of samples is much smaller than the number of genes in microarray gene expression data. Handling this imbalance data as machine learning tasks have the risk of generating an over-fitted learning model, reducing predictability, and unreadability of genetic data. These problems can be significantly decreased by choosing the more informative genes. Unsupervised gene selection techniques can estimate the relation among genes well. Though using mutual information and symmetric uncertainty can estimate the genes' relevancy well, their bivariate measures ignore the possible dependencies among several genes. To address this issue, we propose an unsupervised gene selection scheme based on information theoretic measures. It uses a similarity-based algorithm for gene clustering and then introduces some virtual genes as representatives of gene clusters. These representative genes will have the most common information with the genes in clusters and the least similarity with the representatives of other clusters. The experimental results on benchmark microarray gene expression datasets demonstrate the effectiveness of our approach, as compared to some information theoretic schemes beside to prototype- and density-based clustering methods in both unsupervised and supervised scenarios.

Gene Correlation Guided Gene Selection for Microarray Data Classification.

The microarray cancer data obtained by DNA microarray technology play an important role for cancer prevention, diagnosis, and treatment. However, predicting the different types of tumors is a challenging task since the sample size in microarray data is often small but the dimensionality is very high. Gene selection, which is an effective means, is aimed at mitigating the curse of dimensionality problem and can boost the classification accuracy of microarray data. However, many of previous gene selection methods focus on model design, but neglect the correlation between different genes. In this paper, we introduce a novel unsupervised gene selection method by taking the gene correlation into consideration, named gene correlation guided gene selection (G3CS). Specifically, we calculate the covariance of different gene dimension pairs and embed it into our unsupervised gene selection model to regularize the gene selection coefficient matrix. In such a manner, redundant genes can be effectively excluded. In addition, we utilize a matrix factorization term to exploit the cluster structure of original microarray data to assist the learning process. We design an iterative updating algorithm with convergence guarantee to solve the resultant optimization problem. Experimental results on six publicly available microarray datasets are conducted to validate the efficacy of our proposed method.

Unsupervised gene selection for predicting cell spatial positions in the Drosophila embryo

Analyzing single cell RNA-seq data is important for deciphering the spatial relationships, expression patterns, and developmental processes of cells. Combining in situ hybridization-based gene expression atlas images, some works have successfully recovered spatial locations of cells in zebrafish and Drosophila embryos. In this article, we describe a highly ranked method in the DREAM Single Cell Transcriptomics Challenge for predicting cell positions in the Drosophila embryo. The method performs unsupervised feature extraction to select a small number of driver genes and then uses them to predict gene expression and spatial position of each individual cell. First, hierarchical clustering is used to select a subset of driver genes. Second, the similarity matrix of single cells in the bins of the reference atlas is computed. Based on the similarity matrix, the spatial positions of cells are then determined by hierarchical clustering. This method is evaluated on the cell positions and gene expressions in the DREAM Single Cell Transcriptomics Challenge. The comparison with the “silver standard” suggests that our method is effective in reconstructing the cell spatial positions and gene expression patterns in tissues.

Spatial mapping of single cells in the Drosophila embryo from transcriptomic data based on topological consistency.

The advancement in single-cell RNA sequencing technologies allow us to obtain transcriptome at single cell resolution. However, the original spatial context of cells, a crucial knowledge for understanding cellular and tissue-level functions, is often lost during sequencing. To address this issue, the DREAM Single Cell Transcriptomics Challenge launched a community-wide effort to seek computational solutions for spatial mapping of single cells in tissues using single-cell RNAseq (scRNA-seq) data and a reference atlas obtained from in situ hybridization data. As a top-performing team in this competition, we approach this problem in three steps. The first step involves identifying a set of most informative genes based on the consistency between gene expression similarity and cell proximity. For this step, we propose two different approaches, i.e., an unsupervised approach that does not utilize the gold standard location of the cells provided by the challenge organizers, and a supervised approach that relies on the gold standard locations. In the second step, a Particle Swarm Optimization algorithm is used to optimize the weights of different genes in order to maximize matches between the predicted locations and the gold standard locations. Finally, the information embedded in the cell topology is used to improve the predicted cell-location scores by weighted averaging of scores from neighboring locations. Evaluation results based on DREAM scores show that our method accurately predicts the location of single cells, and the predictions lead to successful recovery of the spatial expression patterns for most of landmark genes. In addition, investigating the selected genes demonstrates that most predictive genes are cluster specific, and stable across our supervised and unsupervised gene selection frameworks. Overall, the promising results obtained by our methods in DREAM challenge demonstrated that topological consistency is a useful concept in identifying marker genes and constructing predictive models for spatial mapping of single cells.

Multi-view feature selection for identifying gene markers: a diversified biological data driven approach

BackgroundIn recent years, to investigate challenging bioinformatics problems, the utilization of multiple genomic and proteomic sources has become immensely popular among researchers. One such issue is feature or gene selection and identifying relevant and non-redundant marker genes from high dimensional gene expression data sets. In that context, designing an efficient feature selection algorithm exploiting knowledge from multiple potential biological resources may be an effective way to understand the spectrum of cancer or other diseases with applications in specific epidemiology for a particular population.ResultsIn the current article, we design the feature selection and marker gene detection as a multi-view multi-objective clustering problem. Regarding that, we propose an Unsupervised Multi-View Multi-Objective clustering-based gene selection approach called UMVMO-select. Three important resources of biological data (gene ontology, protein interaction data, protein sequence) along with gene expression values are collectively utilized to design two different views. UMVMO-select aims to reduce gene space without/minimally compromising the sample classification efficiency and determines relevant and non-redundant gene markers from three cancer gene expression benchmark data sets.ConclusionA thorough comparative analysis has been performed with five clustering and nine existing feature selection methods with respect to several internal and external validity metrics. Obtained results reveal the supremacy of the proposed method. Reported results are also validated through a proper biological significance test and heatmap plotting.

Spatial mapping of single cells in the Drosophila embryo from transcriptomic data based on topological consistency.

The advancement in single-cell RNA sequencing technologies allow us to obtain transcriptome at single cell resolution. However, the original spatial context of cells, a crucial knowledge for understanding cellular and tissue-level functions, is often lost during sequencing. To address this issue, the DREAM Single Cell Transcriptomics Challenge launched a community-wide effort to seek computational solutions for spatial mapping of single cells in tissues using single-cell RNAseq (scRNA-seq) data and a reference atlas obtained from in situ hybridization data. As a top-performing team in this competition, we approach this problem in three steps. The first step involves identifying a set of most informative genes based on the consistency between gene expression similarity and cell proximity. For this step, we propose two different approaches, i.e., an unsupervised approach that does not utilize the gold standard location of the cells provided by the challenge organizers, and a supervised approach that relies on the gold standard locations. In the second step, a Particle Swarm Optimization algorithm is used to optimize the weights of different genes in order to maximize matches between the predicted locations and the gold standard locations. Finally, the information embedded in the cell topology is used to improve the predicted cell-location scores by weighted averaging of scores from neighboring locations. Evaluation results based on DREAM scores show that our method accurately predicts the location of single cells, and the predictions lead to successful recovery of the spatial expression patterns for most of landmark genes. In addition, investigating the selected genes demonstrates that most predictive genes are cluster specific, and stable across our supervised and unsupervised gene selection frameworks. Overall, the promising results obtained by our methods in DREAM challenge demonstrated that topological consistency is a useful concept in identifying marker genes and constructing predictive models for spatial mapping of single cells.

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

Although single-cell RNA sequencing (scRNA-seq) technology is newly invented and a promising one, but because of lack of enough information that labels individual cells, it is hard to interpret the obtained gene expression of each cell. Because of insufficient information available, unsupervised clustering, for example, t-distributed stochastic neighbor embedding and uniform manifold approximation and projection, is usually employed to obtain low-dimensional embedding that can help to understand cell–cell relationship. One possible drawback of this strategy is that the outcome is highly dependent upon genes selected for the usage of clustering. In order to fulfill this requirement, there are many methods that performed unsupervised gene selection. In this study, a tensor decomposition (TD)-based unsupervised feature extraction (FE) was applied to the integration of two scRNA-seq expression profiles that measure human and mouse midbrain development. TD-based unsupervised FE could select not only coincident genes between human and mouse but also biologically reliable genes. Coincidence between two species as well as biological reliability of selected genes is increased compared with that using principal component analysis (PCA)-based FE applied to the same data set in the previous study. Since PCA-based unsupervised FE outperformed the other three popular unsupervised gene selection methods, highly variable genes, bimodal genes, and dpFeature, TD-based unsupervised FE can do so as well. In addition to this, 10 transcription factors (TFs) that might regulate selected genes and might contribute to midbrain development were identified. These 10 TFs, BHLHE40, EGR1, GABPA, IRF3, PPARG, REST, RFX5, STAT3, TCF7L2, and ZBTB33, were previously reported to be related to brain functions and diseases. TD-based unsupervised FE is a promising method to integrate two scRNA-seq profiles effectively.

Cancer Characteristic Gene Selection via Sample Learning Based on Deep Sparse Filtering

Identification of characteristic genes associated with specific biological processes of different cancers could provide insights into the underlying cancer genetics and cancer prognostic assessment. It is of critical importance to select such characteristic genes effectively. In this paper, a novel unsupervised characteristic gene selection method based on sample learning and sparse filtering, Sample Learning based on Deep Sparse Filtering (SLDSF), is proposed. With sample learning, the proposed SLDSF can better represent the gene expression level by the transformed sample space. Most unsupervised characteristic gene selection methods did not consider deep structures, while a multilayer structure may learn more meaningful representations than a single layer, therefore deep sparse filtering is investigated here to implement sample learning in the proposed SLDSF. Experimental studies on several microarray and RNA-Seq datasets demonstrate that the proposed SLDSF is more effective than several representative characteristic gene selection methods (e.g., RGNMF, GNMF, RPCA and PMD) for selecting cancer characteristic genes.

Unsupervised gene selection using biological knowledge : application in sample clustering

BackgroundClassification of biological samples of gene expression data is a basic building block in solving several problems in the field of bioinformatics like cancer and other disease diagnosis and making a proper treatment plan. One big challenge in sample classification is handling large dimensional and redundant gene expression data. To reduce the complexity of handling this high dimensional data, gene/feature selection plays a major role.ResultsThe current paper explores the use of biological knowledge acquired from Gene Ontology database in selecting the proper subset of genes which can further participate in clustering of samples. The proposed feature selection technique is unsupervised in nature as it does not utilize any class label information in the process of gene selection. At the end, a multi-objective clustering approach is deployed to cluster the available set of samples in the reduced gene space.ConclusionsReported results show that consideration of biological knowledge in gene selection technique not only reduces the feature space dimensionality in great extent but also improves the accuracy of sample classification. The obtained reduced gene space is validated using strong biological significance tests. In order to prove the supremacy of our proposed gene selection based sample clustering technique, a thorough comparative analysis has also been performed with state-of-the-art techniques.

Towards Unsupervised Gene Selection: A Matrix Factorization Framework

The recent development of microarray gene expression techniques have made it possible to offer phenotype classification of many diseases. However, in gene expression data analysis, each sample is represented by quite a large number of genes, and many of them are redundant or insignificant to clarify the disease problem. Therefore, how to efficiently select the most useful genes has been becoming one of the most hot research topics in the gene expression data analysis. In this paper, a novel unsupervised two-stage coarse-fine gene selection method is proposed. In the first stage, we apply the kmeans algorithm to over-cluster the genes and discard some redundant genes. In the second stage, we select the most representative genes from the remaining ones based on matrix factorization. Finally the experimental results on several data sets are presented to show the effectiveness of our method.

Unsupervised selection of informative genes in microarray gene expression data

With respect to DNA microarray has produced massive amounts of gene expression data with high dimension in recent years, gene selection is one of the bottlenecks of gene expression datasets analysis. This paper presents a framework for unsupervised gene selection based on dependency maximisation between the samples similarity matrices of before and after deleting a gene, using a novel estimation of the Hilbert-Schmidt independence criterion (HSIC). The key idea is that elimination of genes which are redundant and/or have much relevancy with other genes does not have much effect on pairwise samples similarity. Also, to deal with diagonally dominant matrices, the dynamic range of matrix values is reduced. Additionally, gap statistic and k-means clustering methods are used to increase the speed of proposed methods. Experimental validation is conducted on several microarray gene expression datasets and the results show that our gene selection scheme works well in practice.

Gene selection for microarray data classification using a novel ant colony optimization

The high-dimensionality of microarray data with small number of samples has presented a difficult challenge for the microarray data classification task. The aim of gene selection is to reduce the dimensionality of microarray data in order to enhance the accuracy of the classification task. Existing gene selection methods generally use class labels of the data while due to availability of mislabels or unreliable labels of samples in the microarray data, unsupervised methods could be more essential to the gene selection process. In this paper, we propose an unsupervised gene selection method called MGSACO, which incorporates the ant colony optimization algorithm into the filter approach, by minimizing the redundancy between genes and maximizing the relevance of genes. Moreover, a new fitness function is applied in the proposed method which does not need any learning model to evaluate the subsets of selected genes. Thus, it is classified into the filter approach. The performance of the proposed method is extensively tested upon five publicly available microarray datasets, and it is compared to those of the seven well-known unsupervised and supervised gene selection methods in terms of classification error rates of the three frequently used classifiers including support vector machine, naïve Bayes, and decision tree. Experimental results show that MGSACO is significantly superior to the existing methods over different classifiers and datasets.

A Comparative Analysis of Rough Set Based Intelligent Techniques for Unsupervised Gene Selection

As the micro array databases increases in dimension and results in complexity, identifying the most informative genes is a challenging task. Such difficulty is often related to the huge number of genes with very few samples. Research in medical data mining addresses this problem by applying techniques from data mining and machine learning to the micro array datasets. In this paper Unsupervised Tolerance Rough Set based Quick Reduct (U-TRS-QR), a diverse feature selection algorithm, which extends the existing equivalent rough sets for unsupervised learning, is proposed. Genes selected by the proposed method leads to a considerably improved class predictions in wide experiments on two gene expression datasets: Brain Tumor and Colon Cancer. The results indicate consistent improvement among 12 classifiers.

A composite gene selection for DNA microarray data analysis

An important aspect in microarray data analysis is the selection of an appropriate number of the most relevant genes among a large population of genes. In this study, we have proposed a composite gene selection using both unsupervised and supervised gene selections. In the unsupervised gene selection, we used the threshold number of misclassification (TNoM) score to select an appropriate number of the top-ranked genes for microarray data analysis. In the supervised gene selection, the minimum number of genes showing the highest accuracy is obtained using the non-overlap area distribution measurement (NADM) method provided by the neural network with weighted fuzzy membership functions (NEWFM) from the top-ranked genes. In this study, from a colon cancer dataset and a leukemia dataset, we selected the top-ranked 93 colon cancer and 143 leukemia genes with ≤14 (colon cancer) and ≤13 (leukemia) TNoM scores from a total of 2000 colon cancer and 7129 leukemia genes. By the NADM method, a minimum of 4 colon cancer and 13 leukemia genes were selected from the top-ranked 93 colon cancer and 143 leukemia genes. When the minimal 4 colon cancer and 13 leukemia genes were used as inputs for the NEWFM, the performance accuracies were 98.39 % and 100 % for colon cancer and leukemia, respectively.

Mutual Information-Based Supervised Attribute Clustering for Microarray Sample Classification

Microarray technology is one of the important biotechnological means that allows to record the expression levels of thousands of genes simultaneously within a number of different samples. An important application of microarray gene expression data in functional genomics is to classify samples according to their gene expression profiles. Among the large amount of genes presented in gene expression data, only a small fraction of them is effective for performing a certain diagnostic test. Hence, one of the major tasks with the gene expression data is to find groups of coregulated genes whose collective expression is strongly associated with the sample categories or response variables. In this regard, a new supervised attribute clustering algorithm is proposed to find such groups of genes. It directly incorporates the information of sample categories into the attribute clustering process. A new quantitative measure, based on mutual information, is introduced that incorporates the information of sample categories to measure the similarity between attributes. The proposed supervised attribute clustering algorithm is based on measuring the similarity between attributes using the new quantitative measure, whereby redundancy among the attributes is removed. The clusters are then refined incrementally based on sample categories. The performance of the proposed algorithm is compared with that of existing supervised and unsupervised gene clustering and gene selection algorithms based on the class separability index and the predictive accuracy of naive bayes classifier, K-nearest neighbor rule, and support vector machine on three cancer and two arthritis microarray data sets. The biological significance of the generated clusters is interpreted using the gene ontology. An important finding is that the proposed supervised attribute clustering algorithm is shown to be effective for identifying biologically significant gene clusters with excellent predictive capability.

Fuzzy–Rough Supervised Attribute Clustering Algorithm and Classification of Microarray Data

One of the major tasks with gene expression data is to find groups of coregulated genes whose collective expression is strongly associated with sample categories. In this regard, a new clustering algorithm, termed as fuzzy-rough supervised attribute clustering (FRSAC), is proposed to find such groups of genes. The proposed algorithm is based on the theory of fuzzy-rough sets, which directly incorporates the information of sample categories into the gene clustering process. A new quantitative measure is introduced based on fuzzy-rough sets that incorporates the information of sample categories to measure the similarity among genes. The proposed algorithm is based on measuring the similarity between genes using the new quantitative measure, whereby redundancy among the genes is removed. The clusters are refined incrementally based on sample categories. The effectiveness of the proposed FRSAC algorithm, along with a comparison with existing supervised and unsupervised gene selection and clustering algorithms, is demonstrated on six cancer and two arthritis data sets based on the class separability index and predictive accuracy of the naive Bayes' classifier, the K-nearest neighbor rule, and the support vector machine.