Analysis Of Single-cell RNA Sequencing Data Research Articles

Mcadet: A feature selection method for fine-resolution single-cell RNA-seq data based on multiple correspondence analysis and community detection.

Single-cell RNA sequencing (scRNA-seq) data analysis faces numerous challenges, including high sparsity, a high-dimensional feature space, and biological noise. These challenges hinder downstream analysis, necessitating the use of feature selection methods to identify informative genes, and reduce data dimensionality. However, existing methods for selecting highly variable genes (HVGs) exhibit limited overlap and inconsistent clustering performance across benchmark datasets. Moreover, these methods often struggle to accurately select HVGs from fine-resolution scRNA-seq datasets and minority cell types, which are more difficult to distinguish, raising concerns about the reliability of their results. To overcome these limitations, we propose a novel feature selection framework for scRNA-seq data called Mcadet. Mcadet integrates Multiple Correspondence Analysis (MCA), graph-based community detection, and a novel statistical testing approach. To assess the effectiveness of Mcadet, we conducted extensive evaluations using both simulated and real-world data, employing unbiased metrics for comparison. Our results demonstrate the superior performance of Mcadet in the selection of HVGs in scenarios involving fine-resolution scRNA-seq datasets and datasets containing minority cell populations. Overall, we demonstrate that Mcadet enhances the reliability of selected HVGs, although the impact of HVG selection on various downstream analyses varies and needs to be further investigated.

Deciphering the PI3K-Akt pathway in lung cancer

This computational study explores the immune landscape of early-stage lung adenocarcinoma through single-cell RNA sequencing data analysis. By leveraging paired single-cell data, key insights were gained into the gene expression profiles of tumor samples, focusing on potential therapeutic targets and pathway interactions involved in cancer progression.Single-cell RNA sequencing data from non-lymphocytic immune cells isolated from stage IA lung adenocarcinoma and adjacent normal tissue were computationally processed. Data quality was assessed using sequence quality control tools, and reads were aligned to the hg38 reference genome. Differential gene expression analysis was conducted to identify upregulated and downregulated features. Reactome pathway analysis complemented with literature review was employed to scrutinize potential relationships and pathway interactions. Correlation analysis highlighted key genes involved in tumor progression, including Thrombospondin, HTR6 (5-hydroxytryptamine receptor 6), and LCK.Distinct gene expression patterns were identified in tumor samples, with strong positive correlations between Thrombospondin, HTR6, and LCK. These genes were found to interact within the PI3K-Akt pathway, suggesting their involvement in tumor proliferation and survival. Pathway analysis further revealed molecular crosstalk between these genes and other metabolic processes, offering potential avenues for therapeutic intervention.This computational analysis of single-cell RNA sequencing data identified key gene expression signatures and pathway interactions in early-stage lung adenocarcinoma. The findings, particularly the roles of Thrombospondin, HTR6, and LCK in the PI3K-Akt pathway, provide new insights into potential therapeutic targets for lung cancer treatment.

Left/right asymmetrically expressed ephrin and Flamingo proteins regulate lateralized axon growth in C. elegans.

While the nervous system of bilaterian animals is mainly left-right (L-R) symmetric at the anatomical level, some molecular and functional L-R asymmetries exist. However, the extent of these molecular asymmetries and their functional consequences remain poorly characterized. C. elegans allows to study L-R asymmetries in the nervous system with single-neuron resolution. We have previously shown that a neural bHLH transcription factor, HLH-16/Olig, is L-R asymmetrically expressed in the AIY neuron lineage and regulates AIY axon projections in a L-R asymmetric manner. Here, by combining a candidate approach and single-cell RNA sequencing data analysis, we identify the ephrin protein EFN-2 and the Flamingo protein FMI-1 as downstream targets of HLH-16 that are L-R asymmetrically expressed in the AIY lineage. We show that EFN-2 and FMI-1 collaborate in the L-R asymmetric regulation of axonal growth. EFN-2 may act via a non-canonical receptor of the L1CAM family, SAX-7. Our study reveals novel molecular L-R asymmetries in the C. elegans nervous system and their functional consequences.

NsDCC: dual-level contrastive clustering with nonuniform sampling for scRNA-seq data analysis.

Dimensionality reduction and clustering are crucial tasks in single-cell RNA sequencing (scRNA-seq) data analysis, treated independently in the current process, hindering their mutual benefits. The latest methods jointly optimize these tasks through deep clustering. However, contrastive learning, with powerful representation capability, can bridge the gap that common deep clustering methods face, which requires pre-defined cluster centers. Therefore, a dual-level contrastive clustering method with nonuniform sampling (nsDCC) is proposed for scRNA-seq data analysis. Dual-level contrastive clustering, which combines instance-level contrast and cluster-level contrast, jointly optimizes dimensionality reduction and clustering. Multi-positive contrastive learning and unit matrix constraint are introduced in instance- and cluster-level contrast, respectively. Furthermore, the attention mechanism is introduced to capture inter-cellular information, which is beneficial for clustering. The nsDCC focuses on important samples at category boundaries and in minority categories by the proposed nearest boundary sparsest density weight assignment algorithm, making it capable of capturing comprehensive characteristics against imbalanced datasets. Experimental results show that nsDCC outperforms the six other state-of-the-art methods on both real and simulated scRNA-seq data, validating its performance on dimensionality reduction and clustering of scRNA-seq data, especially for imbalanced data. Simulation experiments demonstrate that nsDCC is insensitive to "dropout events" in scRNA-seq. Finally, cluster differential expressed gene analysis confirms the meaningfulness of results from nsDCC. In summary, nsDCC is a new way of analyzing and understanding scRNA-seq data.

Integrated analyses reveal IDO1 as a prognostic biomarker coexpressed with PD-1 on tumor-associated macrophages in esophageal squamous cell carcinoma.

Inhibition of indolamine-2,3-dioxygenase 1 (IDO1) has been proposed as a promising strategy for cancer immunotherapy; however, it has failed in clinical trials. Macrophages in the tumor microenvironment (TME) contribute to immune escape and serve as potential therapeutic targets. This study investigated the expression pattern of IDO1 in TME and its impact on prognosis and therapeutic response of patients with esophageal squamous cell carcinoma (ESCC). RNA sequencing data from 95 patients with ESCC from The Cancer Genome Atlas (TCGA) database were used to explore the prognostic value of IDO1. Bioinformatics tools were used to estimate scores for stromal and immune cells in tumour tissues, abundance of eight immune cell types in TME, and sensitivity of chemotherapeutic drugs and immune checkpoint (IC) blockage. The results were validated using digitalized immunohistochemistry and multiplexed immunofluorescence in ESCC tissue samples obtained from our clinical center. TCGA and validation data suggested that high expression of IDO1 was associated with poor patient survival, and IDO1 was an independent prognostic factor. IDO1 expression positively correlated with macrophages in TME and PDCD1 within diverse IC genes. Single-cell RNA sequencing data analysis and multiplexed immunofluorescence verified the coexpression of IDO1 and PD-1 in tumor-associated macrophages (TAMs). Patients with high IDO1 expression showed increased sensitivity to various chemotherapeutic drugs, while were more likely to resist IC blockage. This study identifies IDO1 as an independent prognostic indicator of OS in patients with ESCC, reveals a compelling connection of IDO1, PD-1, and TAMs, and explores the sensitivity of patients with high IDO1 expression to chemotherapeutic drugs and their resistance to IC blockade. These findings open new avenues for potential targets in ESCC immunotherapy.

Abstract B058: Rewiring CD8+ T cell responses to PD-1 immune checkpoint blockade in PDAC via the inhibitory receptor PSGL-1

Abstract Pancreatic ductal adenocarcinoma (PDAC) is an aggressive, poorly immunogenic cancer and non-responsive to all currently available treatments, including immune checkpoint blockade (ICB) therapies. As survival rates remain low (13%), there is an urgent need for new therapeutic options. A hallmark of PDAC is the limited infiltration of anti-tumor cytotoxic CD8+ T cells. Analysis of single-cell RNA-sequencing data of PDAC tumors from treatment naïve patients reveals high expression of P-selectin glycoprotein ligand 1 (PSGL-1) on tumor-infiltrating CD8+ T cells. PSGL-1 is an intrinsic negative regulator of CD8+ T cells that promotes the development of functional exhaustion. We therefore hypothesized that PSGL-1-mediated immune suppression is a critical barrier to effective anti-tumor immunity in PDAC. To test this, we used preclinical models of orthotopic injection PDAC tumor cells into pancreata, recapitulating patient responses to PDAC with limited T cell infiltrates and uncontrolled tumor growth in C57BL/6 (wildtype) control mice and compared anti-tumor immune responses in PSGL-1KO (C57BL/6 background) mice. At 28 days post-injection of KPC.4662 PDAC tumor cells, tumor burden in PSGL-1KO mice was reduced by >50% compared to controls, with 45% of PSGL-1KO animals exhibiting >70% reduction. Total immune infiltration (CD45+) was 2-fold greater in tumors from PSGL-1KO mice, with significant increases in infiltrating CD4+ and CD8+ T cells. Detailed analyses revealed that infiltrating CD8+ T cells from PSGL-1KO mice were less terminally differentiated towards exhaustion. PSGL-1KO PD-1+ CD8+ T cells expressed less TOX expression and fewer cells co-expressed CD38 and CD101 while more expressed CX3CR1 compared to controls. Intratumoral PSGL-1KO CD8+ T cells also retained greater functionality as well as indicated by inflammatory cytokine production (IFNg, TNFa) and Granzyme B expression. Moreover, PSGL-1KO mice showed protection against metastatic disease to the lungs. We find that protection by PSGL-1KO mice is T cell dependent as transient depletion of CD4+ and CD8+ T cells prior to tumor cell implantation results in the loss of PDAC tumor growth inhibition by PSGL-1KO mice. Less-differentiated, proliferation-competent precursors of exhausted CD8+ T cells (TPEX) are essential to PD-1 ICB responsiveness in patients and preclinical models. Given that PSGL-1-deficiency limits terminal differentiation of CD8+ T cells in PDAC, we hypothesized that PSGL-1-deficiency would promote responsiveness to PD-1 ICB, which is ineffective as a monotherapy in PDAC patients and mouse models. While therapeutic treatment with anti-PD-1 had no effect on the tumor growth in the control animals, PD-1 ICB resulted in tumor ablation in 85% of PSGL-1KO animals, demonstrating a profound synergistic effect of PD-1 inhibition in the absence of PSGL-1. From these studies, we conclude that PSGL- 1 is a critical inhibitor anti-tumor T cell immunity in PDAC and represents a novel ICB target with high translational potential for promoting immune responses to PDAC tumors. Citation Format: Jennifer L Hope, Yijuan Zhang, Hannah A. F. Hetrick, Evelyn S. Sanchez Hernandez, Gabriele Romano, Sreeja Roy, Michelle Lin, Ashley B Palete, Swetha Maganti, Dennis C Otero, Katelyn T Byrne, Cosimo Commisso, Linda M Bradley. Rewiring CD8T cell responses to PD-1 immune checkpoint blockade in PDAC via the inhibitory receptor PSGL-1 [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Advances in Pancreatic Cancer Research; 2024 Sep 15-18; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2024;84(17 Suppl_2):Abstract nr B058.

Bone marrow adipogenic lineage precursors are the major regulator of bone resorption in adult mice.

Bone resorption by osteoclasts is a critical step in bone remodeling, a process important for maintaining bone homeostasis and repairing injured bone. We previously identified a bone marrow mesenchymal subpopulation, marrow adipogenic lineage precursors (MALPs), and showed that its production of RANKL stimulates bone resorption in young mice using Adipoq-Cre. To exclude developmental defects and to investigate the role of MALPs-derived RANKL in adult bone, we generated inducible reporter mice (Adipoq-CreER Tomato) and RANKL deficient mice (Adipoq-CreER RANKLflox/flox, iCKO). Single cell-RNA sequencing data analysis, lineage tracing, and in situ hybridization revealed that Adipoq+ cells contain not only MALPs but also late mesenchymal progenitors capable of osteogenic differentiation. However, RANKLmRNA was only detected in MALPs, but not in osteogenic cells. RANKL deficiency in MALPs induced at 3 months of age rapidly increased trabecular bone mass in long bones as well as vertebrae within 1 month due to diminished bone resorption but had no effect on the cortical bone. Ovariectomy (OVX) induced trabecular bone loss at both sites. RANKL depletion either before OVX or at 6 weeks post OVX protected and restored trabecular bone mass. Furthermore, bone healing after drill-hole injury was delayed in iCKO mice. Together, our findings demonstrate that MALPs play a dominant role in controlling trabecular bone resorption and that RANKL from MALPs is essential for trabecular bone turnover in adult bone homeostasis, postmenopausal bone loss, and injury repair.

ScParser: sparse representation learning for scalable single-cell RNA sequencing data analysis

The rapid rise in the availability and scale of scRNA-seq data needs scalable methods for integrative analysis. Though many methods for data integration have been developed, few focus on understanding the heterogeneous effects of biological conditions across different cell populations in integrative analysis. Our proposed scalable approach, scParser, models the heterogeneous effects from biological conditions, which unveils the key mechanisms by which gene expression contributes to phenotypes. Notably, the extended scParser pinpoints biological processes in cell subpopulations that contribute to disease pathogenesis. scParser achieves favorable performance in cell clustering compared to state-of-the-art methods and has a broad and diverse applicability.

EMT-related gene classifications predict the prognosis, immune infiltration, and therapeutic response of osteosarcoma.

Osteosarcoma (OS), a bone tumor with high ability of invasion and metastasis, has seriously affected the health of children and adolescents. Many studies have suggested a connection between OS and the epithelial-mesenchymal transition (EMT). We aimed to integrate EMT-Related genes (EMT-RGs) to predict the prognosis, immune infiltration, and therapeutic response of patients with OS. We used consensus clustering to identify potential EMT-Related OS molecular subtypes. Somatic mutation, tumor immune microenvironment, and functional enrichment analyses were performed for each subtype. We next constructed an EMT-Related risk signature and evaluated it by Kaplan-Meier (K-M) analysis survival and receiver operating characteristic (ROC) curves. Moreover, we constructed a nomogram to more accurately predict OS patients' clinical outcomes. Response effects of immunotherapy in OS patients was analyzed by Tumor Immune Dysfunction and Exclusion (TIDE) analysis, while sensitivity for chemotherapeutic agents was analyzed using oncoPredict. Finally, the expression patterns of hub genes were investigated by single-cell RNA sequencing (scRNA-seq) data analysis. A total of 53 EMT-RDGs related to prognosis were identified, separating OS samples into two separate subgroups. The EMT-high subgroup showed favourable overall survival and more active immune response. Significant correlations were found between EMT-Related DEGs and functions as well as pathways linked to the development of OS. Additionally, a risk signature was established and OS patients were divided into two categories based on the risk scores. The signature presented a good predictive performance and could be recognized as an independent predictive factor for OS. Furthermore, patients with higher risk scores exhibited better sensitivity for five drugs, while no significant difference existed in immunotherapy response between the two risk subgroups. scRNA-seq data analysis displayed different expression patterns of the hub genes. We developed a novel EMT-Related risk signature that can be considered as an independent predictor for OS, which may help improve clinical outcome prediction and guide personalized treatments for patients with OS.

Unraveling the metastatic niche in breast cancer bone metastasis through single-cell RNA sequencing.

Breast cancer (BRCA) is characterized by a unique metastatic pattern, often presenting with bone metastasis (BoM), posing significant clinical challenges. Through the study of the immune microenvironment in BRCA BoM offer perspectives for therapeutic interventions targeting this specific metastatic manifestation of BRCA. This study employs single-cell RNA sequencing and TCGA data analysis to comprehensively compare primary tumors (PT), lymph node metastasis (LN), and BoM. Our investigation identifies a metastatic niche in BoM marked by an increased abundance of cancer-associated fibroblasts (CAFs) and reduced immune cell presence. A distinct subtype (State 1) of BRCA BoM cells associated with adverse prognosis is identified. State 1, displaying heightened stemness traits, may represent an initiation phase for BoM in BRCA. Complex cell communications involving tumor, stromal, and immune cells are revealed. Interactions of FN1, SPP1, and MDK correlate with elevated immune cells in BoM. CD46, MDK, and PTN interactions drive myofibroblast activation and proliferation, contributing to tissue remodeling. Additionally, MDK, PTN, and FN1 interactions influence FAP+ CAF activation, impacting cell adhesion and migration in BoM. These insights deepen our understanding of the metastatic niche in breast cancer BoM.

A disulfidptosis-related lncRNAs cluster to forecast the prognosis and immune landscapes of ovarian cancer.

Disulfidptosis is a newly recognized form of regulated cell death that has been linked to cancer progression and prognosis. Despite this association, the prognostic significance, immunological characteristics and treatment response of disulfidptosis-related lncRNAs (DRLs) in ovarian cancer have not yet been elucidated. The lncRNA data and clinical information for ovarian cancer and normal samples were obtained from the UCSC XENA. Differential expression analysis and Pearson analysis were utilized to identify core DRLs, followed by LASSO algorithm. Random Survival Forest was used to construct a prognostic model. The relationships between risk scores, RNA methylation, immune cell infiltration, mutation, responses to immunotherapy and drug sensitivity analysis were further examined. Additionally, qRT-PCR experiments were conducted to validate the expression of the core DRLs in human ovarian cancer cells and normal ovarian cells and the scRNA-seq data of the core DRLs were obtained from the GEO dataset, available in the TISCH database. A total of 8 core DRLs were obtained to construct a prognostic model for ovarian cancer, categorizing all patients into low-risk and high-risk groups using an optimal cutoff value. The AUC values for 1-year, 3-year and 5-year OS in the TCGA cohort were 0.785, 0.810 and 0.863 respectively, proving a strong predictive capability of the model. The model revealed the high-risk group patients exhibited lower overall survival rates, higher TIDE scores and lower TMB levels compared to the low-risk group. Variations in immune cell infiltration and responses to therapeutic drugs were observed between the high-risk and low-risk groups. Besides, our study verified the correlations between the DRLs and RNA methylation. Additionally, qRT-PCR experiments and single-cell RNA sequencing data analysis were conducted to confirm the significance of the core DRLs at both cellular and scRNA-seq levels. We constructed a reliable and novel prognostic model with a DRLs cluster for ovarian cancer, providing a foundation for further researches in the management of this disease.

Cardiotrophin-1 therapy reduces disease severity in a murine model of glomerular disease.

Cardiotrophin-1 (CT-1), a member of the interleukin (IL)-6 cytokine family, has renoprotective effects in mouse models of acute kidney disease and tubulointerstitial fibrosis, but its role in glomerular disease is unknown. To address this, we used the mouse model of nephrotoxic nephritis to test the hypothesis that CT-1 also has a protective role in immune-mediated glomerular disease. Using immunohistochemistry and analysis of single-cell RNA-sequencing data of isolated glomeruli, we demonstrate that CT-1 is expressed in the glomerulus in male mice, predominantly in parietal epithelial cells and is downregulated in mice with nephrotoxic nephritis. Furthermore, analysis of data from patients revealed that human glomerular disease is also associated with reduced glomerular CT-1 transcript levels. In male mice with nephrotoxic nephritis and established proteinuria, administration of CT-1 resulted in reduced albuminuria, prevented podocyte loss, and sustained plasma creatinine, compared with mice administered saline. CT-1 treatment also reduced fibrosis in the kidney cortex, peri-glomerular macrophage accumulation and the kidney levels of the pro-inflammatory mediator complement component 5a. In conclusion, CT-1 intervention therapy delays the progression of glomerular disease in mice by preserving kidney function and inhibiting renal inflammation and fibrosis.

EPEN-16. BULK AND SINGLE-CELL SPATIAL CHARACTERIZATION OF THE TUMOR IMMUNE MICROENVIRONMENT OF POSTERIOR FOSSA GROUP A EPENDYMOMAS

Abstract BACKGROUND Despite advancements in molecular classification of ependymomas, the tumor immune microenvironment of posterior fossa group A (PFA) ependymomas and the spatial interplay between tumor and immune cells have not been elucidated. Here, we aim to comprehensively characterize the tumor immune microenvironment of pediatric PFA ependymomas, which could ultimately uncover leads for immunotherapy for these children. METHODS We collected bulk RNA-sequencing data of 28 PFA ependymoma samples and defined two subgroups based on the expression of immune-related genes (n = 716). Furthermore, we analyzed single-cell RNA-sequencing data of PFA ependymomas to dissect myeloid and lymphocyte heterogeneity, which we subsequently used to deconvolute bulk data. Lastly, we performed both high-dimensional cyclical immunofluorescence of 20 protein markers on a tissue microarray containing 60 cores from 20 tumor samples and immunohistochemistry on whole tissue slides. RESULTS Unsupervised hierarchical clustering of bulk RNA expression of immune-related genes revealed the presence of an immune-enriched subgroup. Furthermore, analysis of single-cell RNA-sequencing data identified multiple distinct myeloid cell subpopulations, with gene expression signatures reflective of inflammation, angiogenesis/hypoxia response and proliferation. Deconvolution of the bulk RNA-sequencing data using the newly annotated single-cell data as a reference showed enrichment of both myeloid cells and lymphocytes in the immune-enriched subgroup. Lastly, spatial immune profiling enabled detailed mapping of the organization of the tumor immune microenvironment and unveiled a heterogeneous presence of myeloid cells, with enrichment of infiltrating myeloid and T-cells in peri-necrotic areas compared to areas with high tumor cell density. CONCLUSIONS Our multi-faceted approach has shed unprecedented light on the cellular and spatial heterogeneity within the PFA ependymoma immune microenvironment. Further studies are warranted to unravel the functional differences between the two subgroups and the potential impact on response to future immunotherapeutic approaches/interventions.

Augmentation of hepatocellular carcinoma malignancy by annexin A5 through modulation of invasion and angiogenesis

ABSTRUCT Background Hepatocellular carcinoma (HCC) continues to play a substantial role in cancer-related morbidity and mortality, largely owing to its pronounced tumor heterogeneity and propensity for recurrence. This underscores the pressing need for in-depth examination of its highly malignant mechanisms. Annexin A5 (ANXA5), recognized as a hallmark tumor protein, has emerged as a focal point of interest because of its ambiguous function and mechanism in HCC prognosis. This study aimed to provide a comprehensive understanding of the role of ANXA5 in the malignant progression of human HCC cells by employing an integrative approach that combines conventional experimental methods with RNA sequencing. Methods Differences in ANXA5 expression between HCC tissues and corresponding nontumor tissues were evaluated using immunofluorescence (n = 25). Correlation analysis was subsequently performed to assess the association between ANXA5 expression and clinicopathological features (n = 65). The role of ANXA5 in human HCC cell lines with ANXA5 gene knockout and overexpression was explored in vitro using migration and invasion assays and Ki-67 indices and in vivo based on node mice xenograft model. A tube formation assay using human umbilical vein endothelial cells (HUVECs) was conducted to demonstrate the angiogenic effects of ANXA5 in HCC. Single-cell and bulk RNA sequencing was used to further investigate the underlying mechanisms involved. Results This study revealed that ANXA5 is highly expressed in patients with HCC and correlates with poor prognosis. Assays for migration, invasion, and proliferation based on ANXA5 gene knockout and overexpression systems in human HCC cell lines have demonstrated that ANXA5 enhances HCC malignancy in vitro and in vivo. Tube formation assays of HUVECs indicated that ANXA5 facilitates angiogenesis and recruits endothelial cells to HCC cells. Single-cell and bulk RNA sequencing data analysis further confirmed that ANXA5 expression in HCC is associated with hepatocyte metabolism, immune response activation, and various oncogenic signaling pathways. Conclusions This study revealed a meaningful association between elevated ANXA5 expression in tumor tissues and an unfavorable prognosis in patients with HCC. In addition, ANXA5 promotes HCC malignancy by promoting invasion and angiogenesis. Thus, ANXA5 has emerged as a promising therapeutic target for HCC and has the potential to improve patient outcomes.

Cauchy hyper-graph Laplacian nonnegative matrix factorization for single-cell RNA-sequencing data analysis.

Many important biological facts have been found as single-cell RNA sequencing (scRNA-seq) technology has advanced. With the use of this technology, it is now possible to investigate the connections among individual cells, genes, and illnesses. For the analysis of single-cell data, clustering is frequently used. Nevertheless, biological data usually contain a large amount of noise data, and traditional clustering methods are sensitive to noise. However, acquiring higher-order spatial information from the data alone is insufficient. As a result, getting trustworthy clustering findings is challenging. We propose the Cauchy hyper-graph Laplacian non-negative matrix factorization (CHLNMF) as a unique approach to address these issues. In CHLNMF, we replace the measurement based on Euclidean distance in the conventional non-negative matrix factorization (NMF), which can lessen the influence of noise, with the Cauchy loss function (CLF). The model also incorporates the hyper-graph constraint, which takes into account the high-order link among the samples. The CHLNMF model's best solution is then discovered using a half-quadratic optimization approach. Finally, using seven scRNA-seq datasets, we contrast the CHLNMF technique with the other nine top methods. The validity of our technique was established by analysis of the experimental outcomes.

Prognostic model based on B cell marker genes for NSCLC patients under neoadjuvant immunotherapy by integrated analysis of single-cell and bulk RNA-sequencing data.

Neoadjuvant immunotherapy has evolved as an effective option to treat non-small cell lung cancer (NSCLC). B cells play essential roles in the immune system as well as cancer progression. However, the repertoire of B cells and its association with clinical outcomes remains unclear in NSCLC patients receiving neoadjuvant immunotherapy. Single-cell RNA sequencing (scRNA-seq) and bulk RNA sequencing data for LUAD samples were accessed from the TCGA and GEO databases. LUAD-related B cell marker genes were confirmed based on comprehensive analysis of scRNA-seq data. We then constructed the B cell marker gene signature (BCMGS) and validated it. In addition, we evaluated the association of BCGMS with tumor immune microenvironment (TIME) characteristics. Furthermore, we validated the efficacy of BCGMS in a cohort of NSCLC patients receiving neoadjuvant immunotherapy. A BCMGS was constructed based on the TCGA cohort and further validated in three independent GSE cohorts. In addition, the BCMGS was proven to be significantly associated with TIME characteristics. Moreover, a relatively higher risk score indicated poor clinical outcomes and a worse immune response among NSCLC patients receiving neoadjuvant immunotherapy. We constructed an 18-gene prognostic signature derived from B cell marker genes based on scRNA-seq data, which had the potential to predict the prognosis and immune response of NSCLC patients receiving neoadjuvant immunotherapy.

The Workflow for Computational Analysis of Single-cell RNA-sequencing Data

The Workflow for Computational Analysis of Single-cell RNA-sequencing Data

CTEC: a cross-tabulation ensemble clustering approach for single-cell RNA sequencing data analysis.

Cell-type clustering is a crucial first step for single-cell RNA-seq data analysis. However, existing clustering methods often provide different results on cluster assignments with respect to their own data pre-processing, choice of distance metrics, and strategies of feature extraction, thereby limiting their practical applications. We propose Cross-Tabulation Ensemble Clustering (CTEC) method that formulates two re-clustering strategies (distribution- and outlier-based) via cross-tabulation. Benchmarking experiments on five scRNA-Seq datasets illustrate that the proposed CTEC method offers significant improvements over the individual clustering methods. Moreover, CTEC-DB outperforms the state-of-the-art ensemble methods for single-cell data clustering, with 45.4% and 17.1% improvement over the single-cell aggregated from ensemble clustering method (SAFE) and the single-cell aggregated clustering via Mixture model ensemble method (SAME), respectively, on the two-method ensemble test. The source code of the benchmark in this work is available at the GitHub repository https://github.com/LWCHN/CTEC.git.

ScEWE: high-order element-wise weighted ensemble clustering for heterogeneity analysis of single-cell RNA-sequencing data.

With the emergence of large amount of single-cell RNA sequencing (scRNA-seq) data, the exploration of computational methods has become critical in revealing biological mechanisms. Clustering is a representative for deciphering cellular heterogeneity embedded in scRNA-seq data. However, due to the diversity of datasets, none of the existing single-cell clustering methods shows overwhelming performance on all datasets. Weighted ensemble methods are proposed to integrate multiple results to improve heterogeneity analysis performance. These methods are usually weighted by considering the reliability of the base clustering results, ignoring the performance difference of the same base clustering on different cells. In this paper, we propose a high-order element-wise weighting strategy based self-representative ensemble learning framework: scEWE. By assigning different base clustering weights to individual cells, we construct and optimize the consensus matrix in a careful and exquisite way. In addition, we extracted the high-order information between cells, which enhanced the ability to represent the similarity relationship between cells. scEWE is experimentally shown to significantly outperform the state-of-the-art methods, which strongly demonstrates the effectiveness of the method and supports the potential applications in complex single-cell data analytical problems.

Unsupervised Gene-Cell Collective Representation Learning with Optimal Transport

Cell type identification plays a vital role in single-cell RNA sequencing (scRNA-seq) data analysis. Although many deep embedded methods to cluster scRNA-seq data have been proposed, they still fail in elucidating the intrinsic properties of cells and genes. Here, we present a novel end-to-end deep graph clustering model for single-cell transcriptomics data based on unsupervised Gene-Cell Collective representation learning and Optimal Transport (scGCOT) which integrates both cell and gene correlations. Specifically, scGCOT learns the latent embedding of cells and genes simultaneously and reconstructs the cell graph, the gene graph, and the gene expression count matrix. A zero-inflated negative binomial (ZINB) model is estimated via the reconstructed count matrix to capture the essential properties of scRNA-seq data. By leveraging the optimal transport-based joint representation alignment, scGCOT learns the clustering process and the latent representations through a mutually supervised self optimization strategy. Extensive experiments with 14 competing methods on 15 real scRNA-seq datasets demonstrate the competitive edges of scGCOT.