Clustering Of Gene Expression Data Research Articles

Large-Scale Gene Expression Data Clustering through Incremental Ensemble Approach

Abstract DNA microarray technology monitors gene activity in real-time in living organisms. It creates a large amount of data that helps scientists learn about how genes work. Clustering this data helps understand gene interactions and uncover important biological processes. However, the traditional clustering techniques have difficulties due to the enormous dimensionality of gene expression data and the intricacy of biological networks. Although ensemble clustering is a viable strategy, such high-dimensional data may not lend itself well to traditional approaches. This study introduces a novel technique for gene expression data clustering called Incremental Ensemble Clustering for Gene expression data (IECG). There are two steps in the IECG. A technique for grouping gene expression data into windows is presented in the first step, producing a tree of clusters. This procedure is carried out again for succeeding windows that have distinct feature sets. The base clusterings of two consecutive windows are ensembled using a new goal function to form a new clustering solution. By repeating this step by step method for further windows, reliable patterns that are beneficial for medical applications can be extracted. The results from both biological and non-biological data demonstrate that the proposed algorithm outperformed the state-of-the-art algorithms. Additionally, the running time of the proposed algorithm has been examined.

3D clustering of gene expression data from systemic autoinflammatory diseases using self-organizing maps (Clust3D)

Background and objectiveSystemic autoinflammatory diseases (SAIDs) are characterized by widespread inflammation, but for most of them there is a lack of specific biomarkers for accurate diagnosis. Although a number of machine learning algorithms have been used to analyze SAID datasets, aiding in the discovery of novel biomarkers, there is a growing recognition of the importance of SAID timeseries clustering, as it can capture the temporal dynamics of gene expression patterns. MethodologyThis paper proposes a novel clustering methodology to efficiently associate three-dimensional data. The algorithm utilizes competitive learning to create a self-organizing neural network and adjust neuron positions in time-dependent and high dimensional feature space in order to assign them as clustering centers. The quantitative evaluation of the clustering was based on well-known clustering indices. Furthermore, a differential expression analysis and classification pipeline was employed to assess the capability of the proposed methodology to extract more accurate pathway-specific genes from its clusters. For that, a comparative analysis was also conducted against a heuristic timeseries clustering method. ResultsThe proposed methodology achieved better overall clustering indices scores and classification metrics using genes derived from its clusters. Notable cases include a threefold increase in the Calinski-Harabasz clustering index, a twofold improvement in the Davies–Bouldin clustering index and a ∼60% increase in the classification specificity score. ConclusionA novel clustering methodology was developed and applied on several gene expression timeseries datasets from systemic autoinflammatory diseases, and its ability to efficiently produce well separated clusters compared to existing heuristic methods was demonstrated.

Identification of key molecules in the formation of portal vein tumor thrombus in hepatocellular carcinoma based on single cell transcriptomics and in vitro experiments.

The presence of portal vein tumor thrombus (PVTT) is a significant indicator of advanced-stage hepatocellular carcinoma (HCC). Unfortunately, the prediction of PVTT occurrence remains challenging, and there is a lack of comprehensive research exploring the underlying mechanisms of PVTT formation and its association with immune infiltration. Our approach involved analyzing single-cell sequencing data, applying high dimensional weighted gene co-expression network analysis (hdWGCNA), and identifying key genes associated with PVTT development. Furthermore, we constructed competing endogenous RNA (ceRNA) networks and employed weighted gene co-expression network analysis (WGCNA), as well as three machine-learning techniques, to identify the upstream regulatory microRNAs (miRNAs) and long non-coding RNAs (lncRNAs) of the crucial mRNAs. We employed fuzzy clustering of time series gene expression data (Mfuzz), gene set variation analysis (GSVA), and cell communication analysis to uncover significant signaling pathways involved in the activation of these important mRNAs during PVTT development. In addition, we conducted immune infiltration analysis, survival typing, and drug sensitivity analysis using The Cancer Genome Atlas (TCGA) cohort to gain insights into the two patient groups under study. Through the implementation of hdWGCNA, we identified 110 genes that was closely associated with PVTT. Among these genes, TMEM165 emerged as a crucial candidate, and we further investigated its significance using COX regression analysis. Furthermore, through machine learning techniques and survival analysis, we successfully identified the upstream regulatory miRNA (hsa-miR-148a) and lncRNA (LINC00909) that targeted TMEM165. These findings shed light on the complex regulatory network surrounding TMEM165 in the context of PVTT. Moreover, we conducted CIBERSORT analysis, which unveiled correlations between TMEM165 and immune infiltration in HCC patients. Specifically, TMEM165 exhibited associations with various immune cell populations, including memory B cells and CD8+ T cells. Additionally, we observed implications for immune function, particularly in relation to immune checkpoints, within the context of HCC. The regulatory axis involving TMEM165, hsa-miR-148a, and LINC00909 emerges as a crucial determinant in the development of PVTT in HCC patients, and it holds significant implications for prognosis. Furthermore, alterations in the TMEM165/hsa-miR-148a/LINC00909 regulatory axis exhibit a strong correlation with immune infiltration within the HCC tumor microenvironment, leading to immune dysfunction and potential failure of immunotherapy.

Label-aware distance mitigates temporal and spatial variability for clustering and visualization of single-cell gene expression data

Clustering and visualization are essential parts of single-cell gene expression data analysis. The Euclidean distance used in most distance-based methods is not optimal. The batch effect, i.e., the variability among samples gathered from different times, tissues, and patients, introduces large between-group distance and obscures the true identities of cells. To solve this problem, we introduce Label-Aware Distance (Lad), a metric using temporal/spatial locality of the batch effect to control for such factors. We validate Lad on simulated data as well as apply it to a mouse retina development dataset and a lung dataset. We also found the utility of our approach in understanding the progression of the Coronavirus Disease 2019 (COVID-19). Lad provides better cell embedding than state-of-the-art batch correction methods on longitudinal datasets. It can be used in distance-based clustering and visualization methods to combine the power of multiple samples to help make biological findings.

Interdependence of Molecular Lesions That Drive Uveal Melanoma Metastasis.

The metastatic risk of uveal melanoma (UM) is defined by a limited number of molecular lesions, somatic mutations (SF3B1 and BAP1), and copy number alterations (CNA): monosomy of chromosome 3 (M3), chr8q gain (8q), chr6p gain (6p), yet the sequence of events is not clear. We analyzed data from three datasets (TCGA-UVM, GSE27831, GSE51880) with information regarding M3, 8q, 6p, SF3B1, and BAP1 status. We confirm that BAP1 mutations are always associated with M3 in high-risk patients. All other features (6p, 8q, M3, SF3B1 mutation) were present independently from each other. Chr8q gain was frequently associated with chr3 disomy. Hierarchical clustering of gene expression data of samples with different binary combinations of aggressivity factors shows that patients with 8q|M3, BAP1|M3 form one cluster enriched in samples that developed metastases. Patients with 6p combined with either 8q or SF3B1 are mainly represented in the other, low-risk cluster. Several gene expression events that show a non-significant association with outcome when considering single features become significant when analyzing combinations of risk features indicating additive action. The independence of risk factors is consistent with a random risk model of UM metastasis without an obligatory sequence.

Identification of Mild Cognitive Impairment Subtypes using an Interpretable Neural Network based Clustering of Gene Expression Data and Neuroimaging Markers.

In this study, an attempt is made to cluster the gene expression data and neuroimaging markers using an interpretable neural network model to identify Mild Cognitive Impairment (MCI) subtypes. For this, structural Magnetic Resonance (MR) brain images and gene expression data of early and late MCI subjects are considered from a public database. A neural network model is employed to cluster the gene expression data and regional MR volumes. To evaluate the performance of model, clustering metrics are employed and model is explained using perturbation-based method. Results indicate that the developed model is able to identify MCI subtypes. The network learns latent embeddings of disease-specific genes and MR images markers. The clustering metrics are found to be highest when both the imaging and genetic markers are employed. Volumes of lateral ventricles, hippocampus, amygdala and thalamus are found to be associated with late MCI. Significant scores suggest that genes such as StAR, CCDC108, APOO, TRMT13, RASAL2 and ZNF43 play a key role in identifying the MCI subtypes.Clinical Relevance-Identifying distinct MCI subtypes offer potential for precision diagnostics and targeted clinical recruitment.

FGF-trapping hampers cancer stem-like cells in uveal melanoma

BackgroundCancer stem-like cells (CSCs) are a subpopulation of tumor cells responsible for tumor initiation, metastasis, chemoresistance, and relapse. Recently, CSCs have been identified in Uveal Melanoma (UM), which represents the most common primary tumor of the eye. UM is highly resistant to systemic chemotherapy and effective therapies aimed at improving overall survival of patients are eagerly required.MethodsHerein, taking advantage from a pan Fibroblast Growth Factor (FGF)-trap molecule, we singled out and analyzed a UM-CSC subset with marked stem-like properties. A hierarchical clustering of gene expression data publicly available on The Cancer Genome Atlas (TCGA) was performed to identify patients’ clusters.ResultsBy disrupting the FGF/FGF receptor (FGFR)-mediated signaling, we unmasked an FGF-sensitive UM population characterized by increased expression of numerous stemness-related transcription factors, enhanced aldehyde dehydrogenase (ALDH) activity, and tumor-sphere formation capacity. Moreover, FGF inhibition deeply affected UM-CSC survival in vivo in a chorioallantoic membrane (CAM) tumor graft assay, resulting in the reduction of tumor growth. At clinical level, hierarchical clustering of TCGA gene expression data revealed a strong correlation between FGFs/FGFRs and stemness-related genes, allowing the identification of three distinct clusters characterized by different clinical outcomes.ConclusionsOur findings support the evidence that the FGF/FGFR axis represents a master regulator of cancer stemness in primary UM tumors and point to anti-FGF treatments as a novel therapeutic strategy to hit the CSC component in UM.

Finite mixtures of matrix variate Poisson-log normal distributions for three-way count data.

Three-way data structures, characterized by three entities, the units, the variables and the occasions, are frequent in biological studies. In RNA sequencing, three-way data structures are obtained when high-throughput transcriptome sequencing data are collected for n genes across p conditions at r occasions. Matrix variate distributions offer a natural way to model three-way data and mixtures of matrix variate distributions can be used to cluster three-way data. Clustering of gene expression data is carried out as means of discovering gene co-expression networks. In this work, a mixture of matrix variate Poisson-log normal distributions is proposed for clustering read counts from RNA sequencing. By considering the matrix variate structure, full information on the conditions and occasions of the RNA sequencing dataset is simultaneously considered, and the number of covariance parameters to be estimated is reduced. We propose three different frameworks for parameter estimation: a Markov chain Monte Carlo-based approach, a variational Gaussian approximation-based approach, and a hybrid approach. Various information criteria are used for model selection. The models are applied to both real and simulated data, and we demonstrate that the proposed approaches can recover the underlying cluster structure in both cases. In simulation studies where the true model parameters are known, our proposed approach shows good parameter recovery. The GitHub R package for this work is available at https://github.com/anjalisilva/mixMVPLN and is released under the open source MIT license.

Machine Learning Methodologies for Clustering Gene Expression Data in Cancer

Gene expression data hide vital information required to understand the biological process that takes place in a particular organism. Extracting the hidden patterns in gene expression data helps to strengthen the understanding of functional genomics. The complexity of biological networks and the volume of genes present increase the challenges of comprehending and interpretation of the resulting mass of data, which consists of millions of measurements; these data also inhibit vagueness, imprecision, and noise. Therefore, thousands of gens can be analyzed at a time using clustering techniques is a first step toward addressing these challenges, which is essential in the data mining process to understand natural structures and identify interesting patterns in the underlying gene expression data [2]. The clustering of gene expression data has been proven to be useful in making known the natural structure inherent in gene expression data, understanding gene functions, cellular processes, and subtypes of cells, finding useful information from noisy data, and understanding gene regulation. The other benefit of clustering gene expression data is the identification of homology, which is very important in drug design. Clustering is a useful method that groups items based on certain similarity measures for understanding the structures, functions, regulation of genes, and cellular processes obtained from gene expression data and providing more insight on a given data set [13].
 This review examines the various clustering algorithms applicable to the gene expression data in order to discover and provide useful knowledge of the appropriate clustering technique that will guarantee stability and high degree of accuracy in its analysis procedure.

Gene Expression Data Analysis Using Feature Weighted Robust Fuzzy c-Means Clustering.

Clustering of gene expression data has been proven to be very useful in various applications, i.e., identifying the natural structure inherent in gene expression, understanding gene functions, mining relevant information from noisy data, and understanding gene regulation. In all these applications, genes, i.e., features, play a crucial role in characterizing them into different groups. These features may be relevant, irrelevant, or redundant, but they have different contributions during the clustering process. This paper presents a novel approach by considering the effect of features during the clustering process. In the proposed method, the fuzzy c-means the objective function is modified using a weighted Euclidean distance between the features with a monotonically decreasing function. The monotonically decreasing function helps control the features' contribution during the clustering process to partition the data into more relevant clusters. The proposed approach is validated, and performance is presented in various clustering performance measures on the different standard datasets. These clustering performance measures have also been compared with multiple state-of-the-art methods.

Assisted clustering of gene expression data using regulatory data from partially overlapping sets of individuals

As omics measurements profiled on different molecular layers are interconnected, integrative approaches that incorporate the regulatory effect from multi-level omics data are needed. When the multi-level omics data are from the same individuals, gene expression (GE) clusters can be identified using information from regulators like genetic variants and DNA methylation. When the multi-level omics data are from different individuals, the choice of integration approaches is limited. We developed an approach to improve GE clustering from microarray data by integrating regulatory data from different but partially overlapping sets of individuals. We achieve this through (1) decomposing gene expression into the regulated component and the other component that is not regulated by measured factors, (2) optimizing the clustering goodness-of-fit objective function. We do not require the availability of different omics measurements on all individuals. A certain amount of individual overlap between GE data and the regulatory data is adequate for modeling the regulation, thus improving GE clustering. A simulation study shows that the performance of the proposed approach depends on the strength of the GE-regulator relationship, degree of missingness, data dimensionality, sample size, and the number of clusters. Across the various simulation settings, the proposed method shows competitive performance in terms of accuracy compared to the alternative K-means clustering method, especially when the clustering structure is due mostly to the regulated component, rather than the unregulated component. We further validate the approach with an application to 8,902 Framingham Heart Study participants with data on up to 17,873 genes and regulation information of DNA methylation and genotype from different but partially overlapping sets of participants. We identify clustering structures of genes associated with pulmonary function while incorporating the predicted regulation effect from the measured regulators. We further investigate the over-representation of these GE clusters in pathways of other diseases that may be related to lung function and respiratory health. We propose a novel approach for clustering GE with the assistance of regulatory data that allowed for different but partially overlapping sets of individuals to be included in different omics data.

Unsupervised Grouped Axial Data Modeling via Hierarchical Bayesian Nonparametric Models With Watson Distributions.

This paper aims at proposing an unsupervised hierarchical nonparametric Bayesian framework for modeling axial data (i.e., observations are axes of direction) that can be partitioned into multiple groups, where each observation within a group is sampled from a mixture of Watson distributions with an infinite number of components that are allowed to be shared across different groups. First, we propose a hierarchical nonparametric Bayesian model for modeling grouped axial data based on the hierarchical Pitman-Yor process mixture model of Watson distributions. Then, we demonstrate that by setting the discount parameters of the proposed model to 0, another hierarchical nonparametric Bayesian model based on hierarchical Dirichlet process can be derived for modeling axial data. To learn the proposed models, we systematically develop a closed-form optimization algorithm based on the collapsed variational Bayes (CVB) inference. Furthermore, to ensure the convergence of the proposed learning algorithm, an annealing mechanism is introduced to the framework of CVB inference, leading to an averaged collapsed variational Bayes inference strategy. The merits of the proposed models for modeling grouped axial data are demonstrated through experiments on both synthetic data and real-world applications involving gene expression data clustering and depth image analysis.

Clustering Gene Expressions Using the Table Invitation Prior.

A prior for Bayesian nonparametric clustering called the Table Invitation Prior (TIP) is used to cluster gene expression data. TIP uses information concerning the pairwise distances between subjects (e.g., gene expression samples) and automatically estimates the number of clusters. TIP's hyperparameters are estimated using a univariate multiple change point detection algorithm with respect to the subject distances, and thus TIP does not require an analyst's intervention for estimating hyperparameters. A Gibbs sampling algorithm is provided, and TIP is used in conjunction with a Normal-Inverse-Wishart likelihood to cluster 801 gene expression samples, each of which belongs to one of five different types of cancer.

P614: SUBCLONAL ARCHITECTURE OF CHROMOSOMES REVEALED BY SINGLE-CELL ANALYSIS OF GENE EXPRESSION IN A PATIENT WITH CLONAL EVOLUTION OF RELAPSING/REFRACTORY CLL

Background: Chronic lymphocytic leukemia (CLL) manifests by remarkable intraclonal heterogeneity of genomic defects, especially in patients with relapsed/refractory disease, often connected with complex karyotype. In cancer cells, chromosomal changes are often multiple and complex. Aberrations can be present only in subclones and often underlie disease refractoriness. Using the bulk analysis such as WGS, WES, or genomic array to precisely determine the co-occurrence of aberrations in the individual cell provides limited information. Aims: To perform an in-depth, single-cell RNA sequencing (scRNAseq) study of subclonal changes associated with relapsed/refractory CLL. Methods: A bulk analysis of DNA from separated malignant cells using comprehensive NGS panel LYNX (PMID 34082072) was performed on a representative CLL case at three time-points (TPs): at dg (TP1, month 0, Rai II), before first therapy (TP2, month 19) and during relapse (TP3, month 59). Simultaneously, the transcriptome was analyzed using scRNAseq (Chromium system, 10x Genomics) in 2330 cells (TP1), 2397 cells (TP2), and 1902 cells (TP3). A detailed data analysis was carried out with the Seurat R package. In addition, we used the InferCNV tool to detect the chromosomal aberrations in every tested cell. We identified multiple chromosomal changes and compared the findings with the results of bulk DNA analysis using a genomic array (ThermoFisher Scientific) and FISH. Results: We detected a number of aberrations that were already present at TP1, including del13q, del11q, mutations in NOTCH1, RPS15, ZMYM3, and minor mutations in TP53 and ATM. At TP2, additional changes comprised mutations in PIM1, XPO1, and del(3q). Between TP2 and TP3, the patient underwent several treatment lines (FCR, BR, COP) without an apparent therapeutic effect. The cells from TP3 carried complex changes, including additional mutations in BIRC3 and RB1. Unsupervised clustering of scRNAseq gene expression data defined major clones bearing aberrations (losses on chromosomes 1, 9, 13, 17, 18). Moreover, these clones were composed of cells with additional aberrations defining a number of separate subclones. The proportion of subclones was typically under the detection limit of the genomic array and showed extreme intraclonal heterogeneity of the relapsed/refractory CLL case. Summary/Conclusion: We showcase how genomic alterations influence the expression of affected genes. Such fact can be exploited to reconstruct chromosomal disruptions in individual cells using scRNAseq and to uncover the subclonal architecture of the disease. Grants: MH-CZ_AZV_NU20-08-00314, MH-CZ_AZV_NV19-03-00091, MEYS-CZ_MUNI/A/1330/2021, MH-CZ_RVO_65269705.

A Gene Expression Clustering Method to Extraction of Cell-to-Cell Biological Communication

Graph-based clustering identification is a practical method to detect the communication between nodes in complex networks that has obtained considerable comments. Since identifying different communities in large-scale data is a challenging task, by understanding the communication between the behaviors of the elements in a community (a cluster), the general characteristics of clusters can be predicted. Graph-based clustering methods have played an important role in clustering gene expression data because of their ability to show the relations between the data. In order to be able to identify genes that lead to the development of diseases, the communication between the cells must be established. The communication between different cells can be indicated by the expression of different genes within them. In this study, the problem of cell-to-cell communication is expressed as a graph and the communication are extracted by recognizing the communities. The FANTOM5 dataset is used to simulate and calculate the similarity between cells. After preprocessing and normalizing the data, to convert this data into graphs, the expression of genes in different cells was examined and by considering a threshold and Wilcoxon test, the communication between them were identified through using clustering.

Fuzzy C means Based Evaluation Algorithms For Cancer Gene Expression Data Clustering

The influx of data in bioinformatics is primarily in the form of DNA, RNA, and protein sequences. This condition places a significant burden on scientists and computers. Some genomics studies depend on clustering techniques to group similarly expressed genes into one cluster. Clustering is a type of unsupervised learning that can be used to divide unknown cluster data into clusters. The k-means and fuzzy c-means (FCM) algorithms are examples of algorithms that can be used for clustering. Consequently, clustering is a common approach that divides an input space into several homogeneous zones; it can be achieved using a variety of algorithms. This study used three models to cluster a brain tumor dataset. The first model uses FCM, which is used to cluster genes. FCM allows an object to belong to two or more clusters with a membership grade between zero and one and the sum of belonging to all clusters of each gene is equal to one. This paradigm is useful when dealing with microarray data. The total time required to implement the first model is 22.2589 s. The second model combines FCM and particle swarm optimization (PSO) to obtain better results. The hybrid algorithm, i.e., FCM–PSO, uses the DB index as objective function. The experimental results show that the proposed hybrid FCM–PSO method is effective. The total time of implementation of this model is 89.6087 s. The third model combines FCM with a genetic algorithm (GA) to obtain better results. This hybrid algorithm also uses the DB index as objective function. The experimental results show that the proposed hybrid FCM–GA method is effective. Its total time of implementation is 50.8021 s. In addition, this study uses cluster validity indexes to determine the best partitioning for the underlying data. Internal validity indexes include the Jaccard, Davies Bouldin, Dunn, Xie–Beni, and silhouette. Meanwhile, external validity indexes include Minkowski, adjusted Rand, and percentage of correctly categorized pairings. Experiments conducted on brain tumor gene expression data demonstrate that the techniques used in this study outperform traditional models in terms of stability and biological significance.

Defining imaging sub-phenotypes of psoriatic arthritis: integrative analysis of imaging data and gene expression in a PsA patient cohort.

To define imaging sub-phenotypes in patients with PsA; determine their association with whole blood gene expression and identify biological pathways characterizing the sub-phenotypes. Fifty-five patients with PsA ready to initiate treatment for active disease were prospectively recruited. We performed musculoskeletal ultrasound assessment of the extent of inflammation in the following domains: synovitis, peritenonitis, tenosynovitis and enthesitis. Peripheral whole blood was profiled with RNAseq, and gene expression data were obtained. First, unsupervised cluster analysis was performed to define imaging sub-phenotypes that reflected the predominant tissue involved. Subsequently, principal component analysis was used to determine the association between imaging-defined sub-phenotypes and peripheral blood gene expression profile. Pathway enrichment analysis was performed to identify underlying mechanisms that characterize individual sub-phenotypes. Cluster analysis revealed three imaging sub-phenotypes: (i) synovitis predominant [n = 31 (56%)]; (ii) enthesitis predominant [n = 13 (24%)]; (iii) peritenonitis predominant [n = 11 (20%)]. The peritenonitis-predominant sub-phenotype had the most severe clinical joint involvement, whereas the enthesitis-predominant sub-phenotype had the highest tender entheseal count. Unsupervised clustering of gene expression data identified three sub-phenotypes that partially overlapped with the imaging sub-phenotypes suggesting biological and clinical relevance of these sub-phenotypes. We therefore characterized enriched differential pathways, which included: immune system (innate system, B cells and neutrophil degranulation), complement system, platelet activation and coagulation function. We identified three sub-phenotypes based on the predominant tissue involved in patients with active PsA. Distinct biological pathways may underlie these imaging sub-phenotypes seen in PsA, suggesting their biological and clinical importance.

Targets and Potential Mechanism of Scutellaria baicalensis in Treatment of Primary Hepatocellular Carcinoma Based on Bioinformatics Analysis.

Objective To analyze the target and potential mechanism of Scutellaria baicalensis (SB) in the treatment of HCC based on bioinformatics, so as to provide suggestions for the diagnosis, treatment, and drug development of hepatocellular carcinoma (HCC). Methods The regulated gene targets of SB were screened by gene expression pattern clustering and differential analysis of gene expression data of HepG2 cells treated with SB at 0 h, 1 h, 3 h, 6 h, 12 h, and 24 h. The module genes related to HCC were identified by the weighted gene coexpression network analysis (WGCNA). KEGG and GO enrichment were used to analyze the molecular function and structure of the module, and GSEA was used to evaluate the different functional pathways between normal people and patients with HCC. Then, the module gene was used for univariate Cox proportional hazard analysis and the least absolute shrinkage and selection operator (LASSO) Cox regression analysis to build a prognostic model. The protein-protein interaction network (PPI) was used to analyze the core genes regulated by SB (CGRSB) of the module, and the survival curve revealed the CGRSB impact on patient survival. The CIBERSORT algorithm combined with correlation analysis to explore the relationship between CGRSB and immune infiltration. Finally, the single-cell sequencing technique was used to analyze the distribution of CGRSB at the cellular level. Results SB could regulate 903 genes, of which 234 were related to the occurrence of HCC. The prognosis model constructed by these genes has a good effect in evaluating the survival of patients. KEGG and GO enrichment analysis showed that the regulation of SB on HCC mainly focused on some cell proliferation, apoptosis, and immune-related functions. GSEA enrichment analysis showed that these functions are related to the occurrence of HCC. A total of 24 CGRSB were obtained after screening, of which 13 were significantly related to survival, and most of them were unfavorable factors for patient survival. The correlation analysis of gene expression showed that most of CGRSB was significantly correlated with T cells, macrophages, and other functions. The results of single-cell analysis showed that the distribution of CGRSB in macrophages was the most. Conclusion SB has high credibility in the treatment of HCC, such as CDK2, AURKB, RRM2, CENPE, ESR1, and PRIM2. These targets can be used as potential biomarkers for clinical diagnosis. The research also shows that the p53 signal pathway, MAPK signal pathway, apoptosis pathway, T cell receptor pathway, and macrophage-mediated tumor immunity play the most important role in the mechanism of SB in treating HCC.

Manifold learning based robust clustering of gene expression data for cancer subtyping

Background:Discovering clinically relevant and statistically significant subtypes of cancer is of paramount importance in precision medicine for customizing the practice of medicine to individual needs. For this purpose gene expression profiles of patients is analysed in an unsupervised manner followed by determining the clinical relevance of the subtypes discovered. However, high dimensional nature of omics data leads to ‘curse of dimensionality’ and results in overlapping subtypes with regard to subtype specific survival curves. None of the methods guarantee separability of patients with respect to subtype specific survival curves. Methods:We propose a robust framework for clustering cancer patients into molecular subtypes by leveraging gene expression data. The framework is based on UMAP and an adaptive to noise robust clustering method OTRIMLE. Survival analysis is performed for the discovered subtypes, and the clustering solution that maximizes the Restricted Life Expectancy difference (RLED) between survival curves for each subtype is chosen. Results:The resulting framework is tested on five datasets of cancer obtained from TCGA, and the results are evaluated by comparing with Neighbourhood based multi-omics (NEMO), Similarity Network Fusion (SNF) and Consensus Clustering (CC). Comparisons show that the proposed framework achieved better separation of subtype specific survival curves. For instance, the minimum separation achieved between any two subtypes for GLIO is 94 days for the proposed framework, while it is about 57, 62 and 38 days respectively for NEMO, SNF and CC. Further, over-representation analysis for each subtype is performed to identify significantly over-represented pathways which can be potential treatment targets. Conclusion:The results advocate that our approach is able to discover subtypes that have better separability in terms of survival, and its significance lies in reducing uncertainty in patient’s expected outcome.

Non-heat-stressed Method to Isolate Hepatic Stellate Cells From Highly Steatotic Tumor-bearing Liver Using CD49a.

Non-heat-stressed Method to Isolate Hepatic Stellate Cells From Highly Steatotic Tumor-bearing Liver Using CD49a.