A Single-Cell Transcriptomic Map of the Human and Mouse Pancreas Reveals Inter- and Intra-cell Population Structure

Tumor-associated Schwann cells as new therapeutic target in non-neurological cancers

Cancer neuroscience, a burgeoning field, investigates the complex interactions between cancer and the nervous system, emphasizing how cancer cells exploit neuronal components for growth and metastasis. Tumor-associated Schwann cells (TASc) have emerged as crucial players in the progression of highly innervated cancers, highlighting the intricate relationship between the tumor microenvironment (TME) and the nervous system. This review concludes how TASc, as the most abundant glial cell in the peripheral nervous system, contribute to tumor growth, metastasis, and the remodeling of the TME. Acting similarly to reactive astrocytes in the central nervous system, TASc are implicated in driving perineural invasion (PNI), a distinctive cancer progression pathway facilitating tumor infiltration and metastasis. These TASc not only contribute indirectly to pain but also promote tumor recurrence and poor prognosis. Intrinsic to their role, TASc exhibit unique gene expression profiles and phenotypic transformations, shifting from myelinating to non-myelinating states, thereby actively participating in metastasis and the remodeling of the tumor microenvironment. Targeting TASc represents a novel and promising therapeutic strategy in non-neurological cancers, offering new avenues for clinical intervention.

Identifying similar populations across independent single cell studies without data integration

Supervised and unsupervised methods have emerged to address the complexity of single cell data analysis in the context of large pools of independent studies. Here, we present ClusterFoldSimilarity (CFS), a novel statistical method design to quantify the similarity between cell groups across any number of independent datasets, without the need for data correction or integration. By bypassing these processes, CFS avoids the introduction of artifacts and loss of information, offering a simple, efficient, and scalable solution. This method match groups of cells that exhibit conserved phenotypes across datasets, including different tissues and species, and in a multimodal scenario, including single-cell RNA-Seq, ATAC-Seq, single-cell proteomics, or, more broadly, data exhibiting differential abundance effects among groups of cells. Additionally, CFS performs feature selection, obtaining cross-dataset markers of the similar phenotypes observed, providing an inherent interpretability of relationships between cell populations. To showcase the effectiveness of our methodology, we generated single-nuclei RNA-Seq data from the motor cortex and spinal cord of adult mice. By using CFS, we identified three distinct sub-populations of astrocytes conserved on both tissues. CFS includes various visualization methods for the interpretation of the similarity scores and similar cell populations.

The role of single cell transcriptomics for efficacy and toxicity profiling of chimeric antigen receptor (CAR) T cell therapies

CAR T cells are genetically modified T cells that target specific epitopes. CAR T cell therapy has proven effective in difficult-to-treat B cell cancers and is now expanding into hematology and solid tumors. To date, approved CAR therapies target only two specific epitopes on cancer cells. Identifying more suitable targets is challenged by the lack of truly cancer-specific structures and the potential for on-target off-tumor toxicity. We analyzed gene expression of potential targets in single-cell data from cancer and healthy tissues. Because safety and efficacy can ultimately only be defined clinically, we selected approved and investigational targets for which clinical trail data are available. We generated atlases using >300,000 cells from 48 patients with follicular lymphoma, multiple myeloma, and B-cell acute lymphoblastic leukemia, and integrated over 3 million cells from 35 healthy tissues, harmonizing datasets from over 300 donors. To contextualize findings, we compared target expression patterns with outcome data from clinical trials, linking target profiles to efficacy and toxicity, and ranked 15 investigational targets based on their similarity to approved ones. Target expression did not significantly correlate with reported clinical toxicities in patients undergoing therapy. This may be attributed to the intricate interplay of patient-specific variables, the limited amount of metadata, and the complexity underlying toxicity. Nevertheless, our study serves as a resource for retrospective and prospective target evaluation to improve the safety and efficacy of CAR therapies.

Revealing a novel Decorin-expressing tumor stromal subset in hepatocellular carcinoma via integrative analysis single-cell RNA sequencing

Hepatocellular carcinoma (HCC) remains a leading cause of cancer-related mortality, emphasizing the need for novel therapeutic strategies. Decorin (DCN), a chondroitin sulfate proteoglycan (CSPG), has been proposed as a tumor suppressor, yet its precise role in HCC and the tumor microenvironment (TME) remains underexplored. Through integrated analyses of bulk RNA and single-cell RNA sequencing datasets, we identified a distinct tumor stromal subset highly expressing DCN and associated chondroitin sulfate (CS) synthases. Our findings revealed that DCN expression is significantly downregulated in HCC tissue, but upregulated in peri-tumor stroma, where it correlates with better prognosis and reduced capsular invasion. Western blot analysis demonstrated that CS-DCN, the glycosylated form of DCN, plays a dominant role in this context. Single-cell clustering analysis identified a unique stromal subset in HCC characterized by elevated expression of DCN, CSPGs, and CS synthases, associated with extracellular matrix (ECM) remodeling and protective barrier functions. A six-gene DCN-associated signature derived from this subset, including DCN, BGN, SRPX, CHSY3, CHST3, and CHPF, was validated as a prognostic marker for HCC. Furthermore, functional assays demonstrated that CS-DCN significantly inhibited HCC cell proliferation and invasion. Our study highlights the critical role of DCN in HCC TME and provides insights into its therapeutic potential. Modulating CSPG pathways, particularly on CS-DCN-expressing stromal cells, may offer a promising approach for improving HCC treatment and patient outcomes.

Ligand-receptor dynamics in heterophily-aware graph neural networks for enhanced cell type prediction from single-cell RNA-seq data

Graph Neural Networks (GNNs) have emerged as powerful tools for analyzing structured data, particularly in domains where relationships and interactions between entities are key. By leveraging the inherent graph structure in datasets, GNNs excel in capturing complex dependencies and patterns that traditional neural networks might miss. This advantage is especially pronounced in the field of computational biology, where the intricate connections between biological entities play a crucial role. In this context, Our work explores the application of GNNs to single-cell RNA sequencing (scRNA-seq) data, a domain characterized by complex and heterogeneous relationships. By extracting ligand-receptor (L-R) associations from LIANA and constructing Cell-Cell association networks with varying edge homophily ratios, based on L-R information, we enhance the biological relevance and accuracy of depicting cellular communication pathways. While standard GNN models like Graph Convolutional Networks (GCN), GraphSAGE, Graph Attention Networks (GAT), and MixHop often assume homophily (similar nodes are more likely to be connected), this assumption does not always hold in biological networks. To address this, we explore advanced graph neural network methods, such asH 2 Graph Convolutional Networks and Gated Bi-Kernel GNNs (GBK-GNN), that are specifically designed to handle heterophilic data. Our study spans across six diverse datasets, enabling a thorough comparison between heterophily-aware GNNs and traditional homophily-assuming models, including Multi-Layer Perceptrons, which disregards graph structure entirely. Our findings highlight the importance of considering data-specific characteristics in GNN applications, demonstrating that heterophily-focused methods can effectively decipher the complex patterns within scRNA-seq data. By integrating multi-omics data, including gene expression profiles and L-R interactions, we pave the way for more accurate and insightful analyses in computational biology, offering a more comprehensive understanding of cellular environments and interactions.

DECEPTICON: a correlation-based strategy for RNA-seq deconvolution inspired by a variation of the Anna Karenina principle

Accurately deconvoluting cellular composition from bulk RNA-seq data is pivotal for understanding the tumor microenvironment and advancing precision medicine. Existing methods often struggle to consistently and accurately quantify cell types across heterogeneous RNA-seq datasets, particularly when ground truths are unavailable. In this study, we introduce DECEPTICON, a deconvolution strategy inspired by the Anna Karenina principle, which postulates that successful outcomes share common traits, while failures are more varied. DECEPTICON selects top-performing methods by leveraging correlations between different strategies and combines them dynamically to enhance performance. Our approach demonstrates superior accuracy in predicting cell-type proportions across multiple tumor datasets, improving correlation by 23.9% and reducing root mean square error by 73.5% compared to the best of 50 analyzed strategies. Applied to The Cancer Genome Atlas (TCGA) datasets for breast carcinoma, cervical squamous cell carcinoma, and lung adenocarcinoma, DECEPTICON-based predictions showed improved differentiation between patient prognoses. This correlation-based strategy offers a reliable, flexible tool for deconvoluting complex transcriptomic data and highlights its potential in refining prognostic assessments in oncology and advancing cancer biology.

Multi-view clustering for single-cell RNA-seq data based on graph fusion

Single-cell RNA sequencing (scRNA-seq) provides transcriptome profiling of individual cells, allowing for in-depth studies of cell heterogeneity at cell resolution. While cell clustering lays the basic foundation of scRNA-seq data analysis, the high-dimensionality and frequent dropout events of the data raise great challenges. Although plenty of dedicated clustering methods have been proposed, they often fail to fully explore the underlying data structure. Here, we introduce scMCGF, a new multi-view clustering algorithm based on graph fusion. It utilizes multi-view data generated from transcriptomic data to learn the consistent and complementary information across different view, ultimately constructing a unified graph matrix for robust cell clustering. Specifically, scMCGF utilizes two-dimensional-reduction methods (principal component analysis and diffusion maps) to capture both linear and non-linear characteristics of the data. Additionally, it calculates a cell-pathway score matrix to incorporate pathway-level information. These three features, along with the pre-processed gene expression data, form the multi-view data. scMCGF iteratively refines the structure of similarity graphs of each view through adaptive learning and learns a unified graph matrix by weighting and fusing the individual similarity graph matrix. The final clustering results are obtained by applying the rank constraint on the Laplacian matrix of the unified graph matrix. Experiments results of 13 real data sets reveal that scMCGF outperforms eight state-of-the-art methods in clustering accuracy and robustness. Furthermore, biological analysis validates that the clustering results of scMCGF provide a reliable foundation for downstream investigations.

Advances and applications in single-cell and spatial genomics

The applications of single-cell and spatial technologies in recent times have revolutionized the present understanding of cellular states and the cellular heterogeneity inherent in complex biological systems. These advancements offer unprecedented resolution in the examination of the functional genomics of individual cells and their spatial context within tissues. In this review, we have comprehensively discussed the historical development and recent progress in the field of single-cell and spatial genomics. We have reviewed the breakthroughs in single-cell multi-omics technologies, spatial genomics methods, and the computational strategies employed toward the analyses of single-cell atlas data. Furthermore, we have highlighted the advances made in constructing cellular atlases and their clinical applications, particularly in the context of disease. Finally, we have discussed the emerging trends, challenges, and opportunities in this rapidly evolving field.

Desmosome mutations impact the tumor microenvironment to promote melanoma proliferation

Desmosomes are transmembrane protein complexes that contribute to cell-cell adhesion in epithelia and other tissues. Here, we report the discovery of frequent genetic alterations in the desmosome in human cancers, with the strongest signal seen in cutaneous melanoma, where desmosomes are mutated in more than 70% of cases. In primary but not metastatic melanoma biopsies, the burden of coding mutations in desmosome genes is associated with a strong reduction in desmosome gene expression. Analysis by spatial transcriptomics and protein immunofluorescence suggests that these decreases in expression occur in keratinocytes in the microenvironment rather than in primary melanoma cells. In further support of a microenvironmental origin, we find that desmosome gene knockdown in keratinocytes yields markedly increased proliferation of adjacent melanoma cells in keratinocyte and melanoma cocultures. Similar increases in melanoma proliferation are observed in media preconditioned with desmosome-deficient keratinocytes. Thus, gradual accumulation of desmosome mutations in neighboring cells may prime melanoma cells for neoplastic transformation.

The scRNA-sequencing landscape of pancreatic ductal adenocarcinoma revealed distinct cell populations associated with tumor initiation and progression

Pancreatic ductal adenocarcinoma (PDAC) stands as a formidable malignancy characterized by its profound lethality. The comprehensive analysis of the transcriptional landscape holds immense significance in understanding PDAC development and exploring novel treatment strategies. However, due to the firm consistency of pancreatic cancer samples, the dissociation of single cells and subsequent sequencing can be challenging. Here, we performed single-cell RNA sequencing (scRNA-seq) on 8 PDAC patients with different lymph node metastasis status. We first identified the crucial role of MMP1 in the transition from normal pancreatic cells to cancer cells. The knockdown of MMP1 in pancreatic cancer cell lines decreased the expression of ductal markers such as SOX9 while the overexpression of MMP1 in hTERT-HPNE increased the expression of ductal markers, suggesting its function of maintaining ductal identity. Secondly, we found a S100A2 + tumor subset which fueled lymph node metastasis in PDAC. The knockdown of S100A2 significantly reduced the motility of pancreatic cancer cell lines in both wound healing and transwell migration assays. While overexpression of S100A2 led to increased migratory capability. Moreover, overexpression of S100A2 in KPC1199, a mouse pancreatic cancer cell line, caused a larger tumor burden in a hemi-spleen injection model of liver metastasis. In addition, epithelial-mesenchymal transition-related genes were decreased by S100A2 knockdown revealed by bulk RNA sequencing. We also identified several pivotal contributors to the pro-tumor microenvironment, notably OMD + fibroblast and CCL2 + macrophage. As a result, our study provides valuable insights for early detection of PDAC and promising therapeutic targets for combatting lymph node metastasis.

Zero-shot evaluation reveals limitations of single-cell foundation models

Foundation models such as scGPT and Geneformer have not been rigorously evaluated in a setting where they are used without any further training (i.e., zero-shot). Understanding the performance of models in zero-shot settings is critical to applications that exclude the ability to fine-tune, such as discovery settings where labels are unknown. Our evaluation of the zero-shot performance of Geneformer and scGPT suggests that, in some cases, these models may face reliability challenges and could be outperformed by simpler methods. Our findings underscore the importance of zero-shot evaluations in development and deployment of foundation models in single-cell research.

Heterogeneity-preserving discriminative feature selection for disease-specific subtype discovery

Disease-specific subtype identification can deepen our understanding of disease progression and pave the way for personalized therapies, given the complexity of disease heterogeneity. Large-scale transcriptomic, proteomic, and imaging datasets create opportunities for discovering subtypes but also pose challenges due to their high dimensionality. To mitigate this, many feature selection methods focus on selecting features that distinguish known diseases or cell states, yet often miss features that preserve heterogeneity and reveal new subtypes. To overcome this gap, we develop Preserving Heterogeneity (PHet), a statistical methodology that employs iterative subsampling and differential analysis of interquartile range, in conjunction with Fisher's method, to identify a small set of features that enhance subtype clustering quality. Here, we show that this method can maintain sample heterogeneity while distinguishing known disease/cell states, with a tendency to outperform previous differential expression and outlier-based methods, indicating its potential to advance our understanding of disease mechanisms and cell differentiation.

scAMZI: attention-based deep autoencoder with zero-inflated layer for clustering scRNA-seq data

BACKGROUND

Clustering scRNA-seq data plays a vital role in scRNA-seq data analysis and downstream analyses. Many computational methods have been proposed and achieved remarkable results. However, there are several limitations of these methods. First, they do not fully exploit cellular features. Second, they are developed based on gene expression information and lack of flexibility in integrating intercellular relationships. Finally, the performance of these methods is affected by dropout event.

RESULTS

We propose a novel deep learning (DL) model based on attention autoencoder and zero-inflated (ZI) layer, namely scAMZI, to cluster scRNA-seq data. scAMZI is mainly composed of SimAM (a Simple, parameter-free Attention Module), autoencoder, ZINB (Zero-Inflated Negative Binomial) model and ZI layer. Based on ZINB model, we introduce autoencoder and SimAM to reduce dimensionality of data and learn feature representations of cells and relationships between cells. Meanwhile, ZI layer is used to handle zero values in the data. We compare the performance of scAMZI with nine methods (three shallow learning algorithms and six state-of-the-art DL-based methods) on fourteen benchmark scRNA-seq datasets of various sizes (from hundreds to tens of thousands of cells) with known cell types. Experimental results demonstrate that scAMZI outperforms competing methods.

CONCLUSIONS

scAMZI outperforms competing methods and can facilitate downstream analyses such as cell annotation, marker gene discovery, and cell trajectory inference. The package of scAMZI is made freely available at https://doi.org/10.5281/zenodo.13131559 .

Dissection of single-cell landscapes for the development of chimeric antigen receptor T cells in Hodgkin lymphoma

ABSTRACT

The success of targeted therapies for hematological malignancies has heralded their potential as both salvage treatment and early treatment lines, reducing the need for high-dose, intensive, and often toxic chemotherapeutic regimens. For young patients with classic Hodgkin lymphoma (cHL), immunotherapies provide the possibility to lessen long-term, treatment-related toxicities. However, suitable therapeutic targets are lacking. By integrating single-cell dissection of the tumor landscape and an in-depth, single-cell-based off-tumor antigen prediction, we identify CD86 as a promising therapeutic target in cHL. CD86 is highly expressed on Hodgkin and Reed-Sternberg cancer cells and cHL-specific tumor-associated macrophages. We reveal CD86-CTLA-4 as a key suppressive pathway in cHL, driving T-cell exhaustion. Cellular therapies targeting CD86 had extraordinary efficacy in vitro and in vivo and were safe in immunocompetent mouse models without compromising bacterial host defense in sepsis models. Our results prove the potential value of anti-CD86 immunotherapies for treating cHL.

Self-Supervised Graph Representation Learning for Single-Cell Classification

Accurately identifying cell types in single-cell RNA sequencing data is critical for understanding cellular differentiation and pathological mechanisms in downstream analysis. As traditional biological approaches are laborious and time-intensive, it is imperative to develop computational biology methods for cell classification. However, it remains a challenge for existing methods to adequately utilize the potential gene expression information within the vast amount of unlabeled cell data, which limits their classification and generalization performance. Therefore, we propose a novel self-supervised graph representation learning framework for single-cell classification, named scSSGC. Specifically, in the pre-training stage of self-supervised learning, multiple K-means clustering tasks conducted on unlabeled cell data are jointly employed for model training, thereby mitigating the issue of limited labeled data. To effectively capture the potential interactions among cells, we introduce a locally augmented graph neural network to enhance the information aggregation capability for nodes with fewer neighbors in the cell graph. A range of benchmark experiments demonstrates that scSSGC outperforms existing state-of-the-art cell classification methods. More importantly, scSSGC provides stable performance when faced with cross-datasets, indicating better generalization ability.

Single Nucleus Multiome Analysis Reveals Early Inflammatory Response to High-Fat Diet in Mouse Pancreatic Islets

In periods of sustained hyper-nutrition, pancreatic β-cells undergo functional compensation through transcriptional upregulation of gene programs driving insulin secretion. This adaptation is essential for maintaining systemic glucose homeostasis and metabolic health. Using single nuclei multiomics, we have mapped the early transcriptional compensation mechanisms in murine islets of Langerhans exposed to high-fat diet (HFD) for one and three weeks. We show that β-cells exhibit the largest transcriptional response to HFD, characterized by early activation of proinflammatory eRegulons and downregulation of β-cell identity genes, particularly in a distinct subset of β-cells. Our observations translate to humans, as we observe an increase in the inflammatory gene signatures in human β-cells in pre-diabetes and diabetes. Collectively, these observations point to cellular cross-talk through proinflammatory signaling as a central and early driver of β-cell dysfunction that limits the compensatory capacity of β-cells, which is closely linked to the development of diabetes.

scHeteroNet: A Heterophily-Aware Graph Neural Network for Accurate Cell Type Annotation and Novel Cell Detection

Single-cell RNA sequencing (scRNA-seq) has unveiled extensive cellular heterogeneity, yet precise cell type annotation and the identification of novel cell populations remain significant challenges. scHeteroNet, a novel graph neural network framework specifically designed to leverage heterophily in scRNA-seq data, is presented. Unlike traditional methods that assume homophily, scHeteroNet captures complex cell-cell interactions by integrating information from both immediate and extended cellular neighborhoods, resulting in highly accurate cell representations. Additionally, scHeteroNet incorporates an innovative novelty propagation mechanism that robustly detects previously uncharacterized cell types. Comprehensive evaluations across diverse scRNA-seq datasets demonstrate that scHeteroNet consistently outperforms state-of-the-art approaches in both cell type classification and novel cell detection. This heterophily-aware approach enhances the ability to uncover cellular diversity, providing deeper insights into complex biological systems and advancing the field of single-cell analysis.

SCITUNA: single-cell data integration tool using network alignment

BACKGROUND

As single-cell genomics experiments increase in complexity and scale, the need to integrate multiple datasets has grown. Such integration enhances cellular feature identification by leveraging larger data volumes. However, batch effects-technical variations arising from differences in labs, times, or protocols-pose a significant challenge. Despite numerous proposed batch correction methods, many still have limitations, such as outputting only dimension-reduced data, relying on computationally intensive models, or resulting in overcorrection for batches with diverse cell type composition.

RESULTS

We introduce a novel method for batch effect correction named SCITUNA, a Single-Cell data Integration Tool Using Network Alignment. We perform evaluations on 39 individual batches from four real datasets and a simulated dataset, which include both scRNA-seq and scATAC-seq datasets, spanning multiple organisms and tissues. A thorough comparison of existing batch correction methods using 13 metrics reveals that SCITUNA outperforms current approaches and is successful at preserving biological signals present in the original data. In particular, SCITUNA shows a better performance than the current methods in all the comparisons except for the multiple batch integration of the lung dataset where the difference is 0.004.

CONCLUSION

SCITUNA effectively removes batch effects while retaining the biological signals present in the data. Our extensive experiments reveal that SCITUNA will be a valuable tool for diverse integration tasks.

scCCTR: An iterative selection-based semi-supervised clustering model for single-cell RNA-seq data

Single-cell RNA sequencing (scRNA-seq) enables the analysis of the genome, transcriptome, and epigenome at the single-cell level, providing a critical tool for understanding cellular heterogeneity and diversity. Cell clustering, a key step in scRNA-seq data analysis, reveals population structure by grouping cells with similar expression patterns. However, due to the high dimensionality and sparsity of scRNA-seq data, the performance of existing clustering algorithms remains suboptimal. In this study, we propose a novel clustering algorithm, scCCTR, which performs semi-supervised classification by guiding a deep learning model through iterative selection of high-confidence cells and labels. The algorithm consists of two main components: an iterative selection module and a semi-supervised classification module. In the iterative selection module, scCCTR progressively selects high-confidence cells that exhibit core group features and iteratively optimizes feature representations, constructing a consensus clustering result throughout the iterations. In the semi-supervised classification module, scCCTR uses the selected core data to train a Transformer neural network, which leverages a multi-head attention mechanism to focus on critical information, thereby achieving higher clustering precision. We compared scCCTR with several established cell clustering methods on real datasets, and the results demonstrate that scCCTR outperforms existing methods in terms of accuracy and effectiveness for both cell clustering and visualization. (The code of scCCTR is free available for academic https://github.com/chenjiejie387/scCCTR).

Blockade of glucagon receptor induces α-cell hypersecretion by hyperaminoacidemia in mice

Blockade of the glucagon receptor (GCGR) has been shown to improve glycemic control. However, this therapeutic approach also brings side effects, such as α-cell hyperplasia and hyperglucagonemia, and the mechanisms underlying these side effects remain elusive. Here, we conduct single-cell transcriptomic sequencing of islets from male GCGR knockout (GCGR-KO) mice. Our analysis confirms the elevated expression of Gcg in GCGR-KO mice, along with enhanced glucagon secretion at single-cell level. Notably, Vgf (nerve growth factor inducible) is specifically upregulated in α cells of GCGR-KO mice. Inhibition of VGF impairs the formation of glucagon immature secretory granules and compromises glucagon maturation, lead to reduced α-cell hypersecretion of glucagon. We further demonstrate that activation of both mTOR-STAT3 and ERK-CREB pathways, induced by elevated circulation amino acids, is responsible for upregulation of Vgf and Gcg expression following glucagon receptor blockade. Thus, our findings elucidate parts of the molecular mechanism underlying hyperglucagonemia in GCGR blockade.

CSsingle: A Unified Tool for Robust Decomposition of Bulk and Spatial Transcriptomic Data Across Diverse Single-Cell References

We introduce CSsingle, a novel method that enhances the decomposition of bulk and spatial transcriptomic (ST) data by addressing key challenges in cellular heterogeneity. CSsingle applies cell size correction using ERCC spike-in controls, enabling it to account for variations in RNA content between cell types and achieve accurate bulk data deconvolution. In addition, it enables fine-scale analysis for ST data, advancing our understanding of tissue architecture and cellular interactions, particularly in complex microenvironments. We provide a unified tool for integrating bulk and ST with scRNA-seq data, advancing the study of complex biological systems and disease processes. The benchmark results demonstrate that CSsingle outperforms existing methods in accuracy and robustness. Validation using more than 700 normal and diseased samples from gastroesophageal tissue reveals the predominant presence of mosaic columnar cells (MCCs), which exhibit a gastric and intestinal mosaic phenotype in Barrett's esophagus and esophageal adenocarcinoma (EAC), in contrast to their very low detectable levels in esophageal squamous cell carcinoma and normal gastroesophageal tissue. We revealed a dynamic relationship between MCCs and squamous cells during immune checkpoint inhibitors (ICI)-based treatment in EAC patients, suggesting MCC expression signatures as predictive and prognostic markers of immunochemotherapy outcomes. Our findings reveal the critical role of MCC in the treatment of EAC and its potential as a biomarker to predict outcomes of immunochemotherapy, providing insight into tumor epithelial plasticity to guide personalized immunotherapeutic strategies.

Neonatal fungi promote lifelong metabolic health through macrophage-dependent β cell development

Loss of early-life microbial diversity is correlated with diabetes, yet mechanisms by which microbes influence disease remain elusive. We report a critical neonatal window in mice when microbiota disruption results in lifelong metabolic consequences stemming from reduced β cell development. We show evidence for the existence of a similar program in humans and identify specific fungi and bacteria that are sufficient for β cell growth. The microbiota also plays an important role in seeding islet-resident macrophages, and macrophage depletion during development reduces β cells. Candida dubliniensis increases β cells in a macrophage-dependent manner through distinctive cell wall composition and reduces murine diabetes incidence. Provision of C. dubliniensis after β cell ablation or antibiotic treatment improves β cell function. These data identify fungi as critical early-life commensals that promote long-term metabolic health.

Heterogeneity-Preserving Discriminative Feature Selection for Disease-Specific Subtype Discovery

The identification of disease-specific subtypes can provide valuable insights into disease progression and potential individualized therapies, important aspects of precision medicine given the complex nature of disease heterogeneity. The advent of high-throughput technologies has enabled the generation and analysis of various molecular data types, such as single-cell RNA-seq, proteomic, and imaging datasets, on a large scale. While these datasets offer opportunities for subtype discovery, they also pose challenges in finding subtype signatures due to their high dimensionality. Feature selection, a key step in the machine learning pipeline, involves selecting signatures that reduce feature size for more efficient downstream computational analysis. Although many existing methods focus on selecting features that differentiate known diseases or cell states, they often struggle to identify features that both preserve heterogeneity and reveal subtypes. To address this, we utilized deep metric learning-based feature embedding to explore the statistical properties of features crucial for preserving heterogeneity. Our analysis indicated that features with a notable difference in interquartile range (IQR) between classes hold important subtype information. Guided by this insight, we developed a statistical method called PHet (Preserving Heterogeneity), which employs iterative subsampling and differential analysis of IQR combined with Fisher's method to identify a small set of features that preserve heterogeneity and enhance subtype clustering quality. Validation on public single-cell RNA-seq and microarray datasets demonstrated PHet's ability to maintain sample heterogeneity while distinguishing known disease/cell states, with a tendency to outperform previous differential expression and outlier-based methods. Furthermore, an analysis of a single-cell RNA-seq dataset from mouse tracheal epithelial cells identified two distinct basal cell subtypes differentiating towards a luminal secretory phenotype using PHet-based features, demonstrating promising results in a real-data application. These results highlight PHet's potential to enhance our understanding of disease mechanisms and cell differentiation, contributing significantly to the field of personalized medicine.

AcImpute: a constraint-enhancing smooth-based approach for imputing single-cell RNA sequencing data

MOTIVATION

Single-cell RNA sequencing (scRNA-seq) provides a powerful tool for studying cellular heterogeneity and complexity. However, dropout events in single-cell RNA-seq data severely hinder the effectiveness and accuracy of downstream analysis. Therefore, data preprocessing with imputation methods is crucial to scRNA-seq analysis.

RESULTS

To address the issue of oversmoothing in smoothing-based imputation methods, the presented AcImpute, an unsupervised method that enhances imputation accuracy by constraining the smoothing weights among cells for genes with different expression levels. Compared with nine other imputation methods in cluster analysis and trajectory inference, the experimental results can demonstrate that AcImpute effectively restores gene expression, preserves inter-cell variability, preventing oversmoothing and improving clustering and trajectory inference performance.

AVAILABILITY AND IMPLEMENTATION

The code is available at https://github.com/Liutto/AcImpute.

scMUG: deep clustering analysis of single-cell RNA-seq data on multiple gene functional modules

Single-cell RNA sequencing (scRNA-seq) has revolutionized our understanding of cellular heterogeneity by providing gene expression data at the single-cell level. Unlike bulk RNA-seq, scRNA-seq allows identification of different cell types within a given tissue, leading to a more nuanced comprehension of cell functions. However, the analysis of scRNA-seq data presents challenges due to its sparsity and high dimensionality. Since bioinformatics plays an important role in the analysis of big data and its utility for the welfare of living beings, it has been widely applied in analyzing scRNA-seq data. To address these challenges, we introduce the scMUG computational pipeline, which incorporates gene functional module information to enhance scRNA-seq clustering analysis. The pipeline includes data preprocessing, cell representation generation, cell-cell similarity matrix construction, and clustering analysis. The scMUG pipeline also introduces a novel similarity measure that combines local density and global distribution in the latent cell representation space. As far as we can tell, this is the first attempt to integrate gene functional associations into scRNA-seq clustering analysis. We curated nine human scRNA-seq datasets to evaluate our scMUG pipeline. With the help of gene functional information and the novel similarity measure, the clustering results from scMUG pipeline present deep insights into functional relationships between gene expression patterns and cellular heterogeneity. In addition, our scMUG pipeline also presents comparable or better clustering performances than other state-of-the-art methods. All source codes of scMUG have been deposited in a GitHub repository with instructions for reproducing all results (https://github.com/degiminnal/scMUG).

TGF-β1-mediated intercellular signaling fuels cooperative cellular invasion

Intratumoral heterogeneity drives cancer progression and influences treatment outcomes. The mechanisms underlying how cellular subpopulations communicate and cooperate to impact progression remain largely unknown. Here, we use collective invasion as a model to deconstruct processes underlying non-small cell lung cancer subpopulation cooperation. We reveal that collectively invading packs consist of heterogeneously cycling and non-cycling subpopulations using distinct pathways. We demonstrate that the follower subpopulation secretes transforming growth factor beta one (TGF-β1) to stimulate divergent subpopulation responses-including proliferation, pack cohesion, and JAG1-dependent invasion-depending on cellular context. While isolated followers maintain proliferation in response to TGF-β1, isolated leaders enter a quiescence-like cellular state. In contrast, leaders within a heterogeneous population sustain proliferation to maintain subpopulation proportions. In vivo, both leader and follower subpopulations are necessary for macro-metastatic disease progression. Taken together, these findings highlight that intercellular communication preserves tumor cell heterogeneity and promotes collective behaviors such as invasion and tumor progression.

WCSGNet: a graph neural network approach using weighted cell-specific networks for cell-type annotation in scRNA-seq

Single-cell RNA sequencing (scRNA-seq) has emerged as a powerful tool for understanding cellular heterogeneity, providing unprecedented resolution in molecular regulation analysis. Existing supervised learning approaches for cell type annotation primarily utilize gene expression profiles from scRNA-seq data. Although some methods incorporated gene interaction network information, they fail to use cell-specific gene association networks. This limitation overlooks the unique gene interaction patterns within individual cells, potentially compromising the accuracy of cell type classification. We introduce WCSGNet, a graph neural network-based algorithm for automatic cell-type annotation that leverages Weighted Cell-Specific Networks (WCSNs). These networks are constructed based on highly variable genes and inherently capture both gene expression patterns and gene association network structure features. Extensive experimental validation demonstrates that WCSGNet consistently achieves superior cell type classification performance, ranking among the top-performing methods while maintaining robust stability across diverse datasets. Notably, WCSGNet exhibits a distinct advantage in handling imbalanced datasets, outperforming existing methods in these challenging scenarios. All datasets and codes for reproducing this work were deposited in a GitHub repository (https://github.com/Yi-ellen/WCSGNet).

scCobra allows contrastive cell embedding learning with domain adaptation for single cell data integration and harmonization

The rapid advancement of single-cell technologies has created an urgent need for effective methods to integrate and harmonize single-cell data. Technical and biological variations across studies complicate data integration, while conventional tools often struggle with reliance on gene expression distribution assumptions and over-correction. Here, we present scCobra, a deep generative neural network designed to overcome these challenges through contrastive learning with domain adaptation. scCobra effectively mitigates batch effects, minimizes over-correction, and ensures biologically meaningful data integration without assuming specific gene expression distributions. It enables online label transfer across datasets with batch effects, allowing continuous integration of new data without retraining. Additionally, scCobra supports batch effect simulation, advanced multi-omic integration, and scalable processing of large datasets. By integrating and harmonizing datasets from similar studies, scCobra expands the available data for investigating specific biological problems, improving cross-study comparability, and revealing insights that may be obscured in isolated datasets.

Identification of mouse and human embryonic pancreatic cells with adult Procr+ progenitor transcriptomic and epigenomic characteristics

Background

The quest to find a progenitor cell in the adult pancreas has driven research in the field for decades. Many potential progenitor cell sources have been reported, but so far this is a matter of debate mainly due to reproducibility issues. The existence of adult Procr+ progenitor cells in mice islets has been recently reported. These were shown to comprise ~1% of islet cells, lack expression of Neurog3 and endocrine hormones, and to be capable of differentiating into all endocrine cell types. However, these findings had limited impact, as further evidence supporting the existence and function of Procr+ progenitors has not emerged.

Methods and findings

We report here an unbiased comparison across mouse and human pancreatic samples, including adult islets and embryonic tissue, to track the existence of Procr+ progenitors originally described based on their global gene expression signature. We could not find Procr+ progenitors on other mouse or human adult pancreatic islet samples. Unexpectedly, our results revealed a transcriptionally close mesothelial cell population in the mouse and human embryonic pancreas. These Procr-like mesothelial cells of the embryonic pancreas share the salient transcriptional and epigenomic features of previously reported Procr+ progenitors found in adult pancreatic islets. Notably, we report here that Procr-like transcriptional signature is gradually established in mesothelial cells during mouse pancreas development from E12.5 to E17.5, which has its largest amount. Further supporting a developmentally relevant role in the human pancreas, we additionally report that a transcriptionally similar population is spontaneously differentiated from human pluripotent stem cells cultured in vitro along the pancreatic lineage.

Conclusions

Our results show that, although the previously reported Procr+ progenitor cell population could not be found in other adult pancreatic islet samples, a mesothelial cell population with a closely related transcriptional signature is present in both the mouse and human embryonic pancreas. Several lines of evidence presented in this work support a developmentally relevant function for these Procr-like mesothelial cells.

Selective epithelial expression of KRASG12D in the Oncopig pancreas drives ductal proliferation and desmoplasia that is accompanied by an immune response

Pancreatic ductal adenocarcinoma (PDAC) remains a formidable challenge in oncology, characterized by a high mortality rate, largely attributable to delayed diagnosis and the intricacies of its tumor microenvironment. Innovations in modeling pancreatic epithelial transformation provide valuable insights into the pathogenesis and potential therapeutic strategies for PDAC. We employed a porcine (Oncopig) model, utilizing the Ad-K8-Cre adenoviral vector, to investigate the effects of variable doses (107 to 1010 pfu) on pancreatic epithelial cells. This vector, the expression from which being driven by a Keratin-8 promoter, will deliver Cre-recombinase specifically to epithelial cells. Intraductal pancreatic injections in transgenic Oncopigs (LSL-KRASG12D-TP53R167H) were performed with histologically based evaluation at 2 months post-injection. Specificity of the adenoviral vector was validated through Keratin-8 expression and Cre-recombinase activity. We confirmed that the Ad-K8-Cre adenoviral vector predominantly targets ductal epithelial cells lining both large and small pancreatic ducts, as evidenced by Keratin 8 and CAM5.2 staining. Higher doses resulted in significant tissue morphology changes, including atrophy, and enlarged lymph nodes. Microscopic examination revealed concentration-dependent proliferation of the ductal epithelium, cellular atypia, metaplasia, and stromal alterations. Transgene expression was confirmed with immunohistochemistry. Desmoplastic responses were evident through vimentin, α-SMA, and Masson's trichrome staining, indicating progressive collagen deposition, particularly at the higher vector doses. Our study suggests a distinct dose-response relationship of Ad-K8-Cre in inducing pancreatic epithelial proliferation and possible neoplasia in an Oncopig model. All doses of the vector induced epithelial proliferation; the higher doses also produced stromal alterations, metaplasia, and possible neoplastic transformation. These findings highlight the potential for site-specific activation of oncogenes in large animal models of epithelial tumors, with the ability to induce stromal alterations reminiscent of human PDAC.

Mechanistic elucidation of human pancreatic acinar development using single-cell transcriptome analysis on a human iPSC differentiation model

Few effective treatments have been developed for intractable pancreatic exocrine disorders due to the lack of suitable disease models using human cells. Pancreatic acinar cells differentiated from human induced pluripotent stem cells (hiPSCs) have the potential to solve this issue. In this study, we aimed to elucidate the developmental mechanisms of pancreatic exocrine acinar lineages to establish a directed differentiation method for pancreatic acinar cells from hiPSCs. hiPSC-derived pancreatic endoderm cells were spontaneously differentiated into both pancreatic exocrine and endocrine tissues by implantation into the renal subcapsular space of NOD/SCID mice. Single-cell RNA-seq analysis of the retrieved grafts confirmed the differentiation of pancreatic acinar lineage cells and identified REG4 as a candidate marker for pancreatic acinar progenitor cells. Furthermore, differential gene expression analysis revealed upregulated pathways, including cAMP-related signals, involved in the differentiation of hiPSC-derived pancreatic acinar lineage cells in vivo, and we found that a cAMP activator, forskolin, facilitates the differentiation from hiPSC-derived pancreatic endoderm into pancreatic acinar progenitor cells in our in vitro differentiation culture. Therefore, this platform contributes to our understanding of the developmental mechanisms of pancreatic acinar lineage cells and the establishment of differentiation methods for acinar cells from hiPSCs.

Single-cell atlas of human pancreatic islet and acinar endothelial cells in health and diabetes

Characterization of the vascular heterogeneity within the pancreas has previously been lacking. Here, we develop strategies to enrich islet-specific endothelial cells (ISECs) and acinar-specific endothelial cells (ASECs) from three human pancreases and corroborate these findings with three published pancreatic datasets. Single-cell RNA sequencing reveals the unique molecular signatures of ISECs, including structural genes COL13A1, ESM1, PLVAP, UNC5B, and LAMA4, angiocrine genes KDR, THBS1, BMPs and CXCR4, and metabolic genes ACE, PASK and F2RL3. ASECs display distinct signatures including GPIHBP1, CCL14, CD74, AQP1, KLF4, and KLF2, which may manage the inflammatory and metabolic needs of the exocrine pancreas. Ligand-receptor analysis suggests ISECs and ASECs interact with LUM+ fibroblasts and RGS5+ pericytes and smooth muscle cells via VEGF-A:VEGFR2, CXCL12:CXCR4, and LIF:LIFR pathways. Comparative expression and immunohistochemistry indicate disruption of endothelial-expressed CD74, ESM1, PLVAP, THBD, VWA1, and VEGF-A cross-talk among vascular and other cell types in diabetes. Thus, our data provide a single-cell vascular atlas of human pancreas, enabling deeper understanding of pancreatic pathophysiology in health and disease.

scHNTL: single-cell RNA-seq data clustering augmented by high-order neighbors and triplet loss

MOTIVATION

The rapid development of single-cell RNA sequencing (scRNA-seq) has significantly advanced biomedical research. Clustering analysis, crucial for scRNA-seq data, faces challenges including data sparsity, high dimensionality, and variable gene expressions. Better low-dimensional embeddings for these complex data should maintain intrinsic information while making similar data close and dissimilar data distant. However, existing methods utilizing neural networks typically focus on minimizing reconstruction loss and maintaining similarity in embeddings of directly related cells, but fail to consider dissimilarity, thus lacking separability and limiting the performance of clustering.

RESULTS

We propose a novel clustering algorithm, called scHNTL (scRNA-seq data clustering augmented by high-order neighbors and triplet loss). It first constructs an auxiliary similarity graph and uses a Graph Attentional Autoencoder to learn initial embeddings of cells. Then it identifies similar and dissimilar cells by exploring high-order structures of the similarity graph and exploits a triplet loss of contrastive learning, to improve the embeddings in preserving structural information by separating dissimilar pairs. Finally, this improvement for embedding and the target of clustering are fused in a self-optimizing clustering framework to obtain the clusters. Experimental evaluations on 16 real-world datasets demonstrate the superiority of scHNTL in clustering over the state-of-the-arts single-cell clustering algorithms.

AVAILABILITY AND IMPLEMENTATION

Python implementation of scHNTL is available at Figshare (https://doi.org/10.6084/m9.figshare.27001090) and Github (https://github.com/SWJTU-ML/scHNTL-code).

Biphasic glucose-stimulated insulin secretion over decades: a journey from measurements and modeling to mechanistic insights

Glucose-stimulated insulin release from pancreatic β-cells is critical for maintaining blood glucose homeostasis. An abrupt increase in blood glucose concentration evokes a rapid and transient rise in insulin secretion followed by a prolonged, slower phase. A diminished first phase is one of the earliest indicators of β-cell dysfunction in individuals predisposed to develop type 2 diabetes. Consequently, researchers have explored the underlying mechanisms for decades, starting with plasma insulin measurements under physiological conditions and advancing to single-vesicle exocytosis measurements in individual β-cells combined with molecular manipulations. Based on a chain of evidence gathered from genetic manipulation to in vivo mouse phenotyping, a widely accepted theory posits that distinct functional insulin vesicle pools in β-cells regulate biphasic glucose-stimulated insulin secretion (GSIS) via activation of different metabolic signal pathways. Recently, we developed a high-resolution imaging technique to visualize single vesicle exocytosis from β-cells within an intact islet. Our findings reveal that β-cells within the islet exhibit heterogeneity in their secretory capabilities, which also differs from the heterogeneous Ca2+ signals observed in islet β-cells in response to glucose stimulation. Most importantly, we demonstrate that biphasic GSIS emerges from the interactions among α-, β-, and δ-cells within the islet and is driven by a small subset of hypersecretory β-cells. Finally, we propose that a shift from reductionism to holism may be required to fully understand the etiology of complex diseases such as diabetes.

Inhibition of CD226 co-stimulation suppresses diabetes development in the NOD mouse by augmenting regulatory T cells and diminishing effector T cell function

AIMS/HYPOTHESIS

Immunotherapeutics targeting T cells are crucial for inhibiting autoimmune disease progression proximal to disease onset in type 1 diabetes. There is an outstanding need to augment the durability and effectiveness of T cell targeting therapies by directly restraining proinflammatory T cell subsets, while simultaneously augmenting regulatory T cell (Treg) activity. Here, we present a novel strategy for preventing diabetes incidence in the NOD mouse model using a blocking monoclonal antibody targeting the type 1 diabetes risk-associated T cell co-stimulatory receptor, CD226.

METHODS

Female NOD mice were treated with anti-CD226 at 7-8 weeks of age and then monitored for diabetes incidence and therapeutic mechanism of action.

RESULTS

Compared with isotype-treated controls, anti-CD226-treated NOD mice showed reduced insulitis severity (0.84-fold, p=0.0002) at 12 weeks and decreased disease incidence (HR 0.41, p=0.015) at 30 weeks. Flow cytometric analysis performed 5 weeks post treatment demonstrated reduced proliferation of conventional CD4+ T cells (0.87-fold, p=0.030) and CD8+ (0.78-fold, p=0.0018) effector memory T cells in spleens of anti-CD226-treated mice. Phenotyping of pancreatic Tregs revealed increased CD25 expression (2.05-fold, p=0.0073) and signal transducer and activator of transcription 5 (STAT5) phosphorylation (1.39-fold, p=0.0007) following anti-CD226, with splenic Tregs displaying augmented suppression of CD4+ responder T cells (Tresps) (1.49-fold, p=0.0008, 1:2 Treg:Tresp) in vitro. Anti-CD226-treated mice exhibited reduced frequencies of islet-specific glucose-6-phosphatase catalytic subunit-related protein (IGRP)-reactive CD8+ T cells in the pancreas, using both ex vivo tetramer staining (0.50-fold, p=0.0317) and single-cell T cell receptor sequencing (0.61-fold, p=0.022) approaches. 51Cr-release assays demonstrated reduced cell-mediated lysis of beta cells (0.61-fold, p<0.0001, 1:1 effector:target) by anti-CD226-treated autoreactive cytotoxic T lymphocytes.

CONCLUSIONS/INTERPRETATION

CD226 blockade reduces T cell cytotoxicity and improves Treg function, representing a targeted and rational approach for restoring immune regulation in type 1 diabetes.

Revealing the biological features of the axolotl pancreas as a new research model

Introduction

The pancreas plays a crucial role in digestion and blood glucose regulation. Current animal models, primarily mice and zebrafish, have limited the exploration of pancreatic biology from an evolutionary-developmental perspective. Tetrapod vertebrate axolotl (Ambystoma mexicanum) serves as a valuable model in developmental, regenerative, and evolutionary biology. However, the fundamental biology of the axolotl pancreas remains underexplored. This study aims to characterize the unique developmental, functional, and evolutionary features of the axolotl pancreas to expand the understanding of pancreatic biology.

Methods

We conducted morphological, histological, and transcriptomic analyses to investigate the axolotl pancreas. Pancreatic development was observed using in situ hybridization and immunostaining for key pancreatic markers. RNA sequencing was performed to profile global gene expression during larva and adult stages. And differential gene expression analysis was used to characterize the conserved and unique gene patterns in the axolotl pancreas. Functional assays, including glucose tolerance tests and insulin tolerance tests, were optimized for individual axolotls. To assess pancreatic gene function, Pdx1 mutants were generated using CRISPR/Cas9-mediated gene editing, and their effects on pancreatic morphology, endocrine cell populations, and glucose homeostasis were analyzed.

Results

The axolotl pancreas contains all known pancreatic cell types and develops from dorsal and ventral buds. Both of buds contribute to exocrine and endocrine glands. The dorsal bud produces the major endocrine cell types, while the ventral bud generates α and δ cells, but not β cells. Differential gene expression analysis indicated a transition in global gene expression from pancreatic cell fate commitment and the cell cycle to glucose response, hormone synthesis, and secretion, following the development progression. Notably, the adult axolotl pancreas exhibits slower metabolic activity compared to mammals, as evidenced by the results of GTT and ITT. The mutation of Pdx1 resulted in hyperglycemia and a significant reduction in pancreatic cell mass, including a complete loss of endocrine cells, although it did not lead to a lethal phenotype.

Discussion

This study examines the axolotl pancreas, highlighting the conservation of pancreatic development. Our study highlights the unique features of the axolotl pancreas and broadens the scope of animal models available for pancreatic evolution and disease research.

MetaQ: fast, scalable and accurate metacell inference via single-cell quantization

To overcome the computational barriers of analyzing large-scale single-cell sequencing data, we introduce MetaQ, a metacell algorithm that scales to arbitrarily large datasets with linear runtime and constant memory usage. Inspired by cellular development, MetaQ conceptualizes each metacell as a collective ancestor of biologically similar cells. By quantizing cells into a discrete codebook, where each entry represents a metacell capable of reconstructing the original cells it quantizes, MetaQ identifies homogeneous cell subsets for efficient and accurate metacell inference. This approach reduces computational complexity from exponential to linear while maintaining or surpassing the performance of existing metacell algorithms. Extensive experiments demonstrate that MetaQ excels in downstream tasks such as cell type annotation, developmental trajectory inference, batch integration, and differential expression analysis. Thanks to its superior efficiency and effectiveness, MetaQ makes analyzing datasets with millions of cells practical, offering a powerful solution for single-cell studies in the era of high-throughput profiling.

Inhibition of vascular smooth muscle cell PERK/ATF4 ER stress signaling protects against abdominal aortic aneurysms

Abdominal aortic aneurysms (AAA) are a life-threatening cardiovascular disease for which there is a lack of effective therapy preventing aortic rupture. During AAA formation, pathological vascular remodeling is driven by vascular smooth muscle cell (VSMC) dysfunction and apoptosis, for which the mechanisms regulating loss of VSMCs within the aortic wall remain poorly defined. Using single-cell RNA-Seq of human AAA tissues, we identified increased activation of the endoplasmic reticulum stress response pathway, PERK/eIF2α/ATF4, in aortic VSMCs resulting in upregulation of an apoptotic cellular response. Mechanistically, we reported that aberrant TNF-α activity within the aortic wall induces VSMC ATF4 activation through the PERK endoplasmic reticulum stress response, resulting in progressive apoptosis. In vivo targeted inhibition of the PERK pathway, with VSMC-specific genetic depletion (Eif2ak3fl/fl Myh11-CreERT2) or pharmacological inhibition in the elastase and angiotensin II-induced AAA model preserved VSMC function, decreased elastin fragmentation, attenuated VSMC apoptosis, and markedly reduced AAA expansion. Together, our findings suggest that cell-specific pharmacologic therapy targeting the PERK/eIF2α/ATF4 pathway in VSMCs may be an effective intervention to prevent AAA expansion.

Consensus representation of multiple cell-cell graphs from gene signaling pathways for cell type annotation

BACKGROUND

Recent advancements in single-cell RNA sequencing have greatly expanded our knowledge of the heterogeneous nature of tissues. However, robust and accurate cell type annotation continues to be a major challenge, hindered by issues such as marker specificity, batch effects, and a lack of comprehensive spatial and interaction data. Traditional annotation methods often fail to adequately address the complexity of cellular interactions and gene regulatory networks.

RESULTS

We proposed scMCGraph, a comprehensive computational framework that integrates gene expression with pathway activity to accurately annotate cell types within diverse scRNA-seq datasets. Initially, our model constructs multiple pathway-specific views using various pathway databases, which reflect both gene expression and pathway activities. These pathway-specific views are then integrated into a consensus graph. The consensus graph is subsequently utilized to reconstruct the multiple pathway views. Our model demonstrated exceptional robustness and accuracy across various analyses, including cross-platform, cross-time, cross-sample, and clinical dataset evaluations.

CONCLUSIONS

scMCGraph represents a significant advance in cell type annotation. The experiments have demonstrated that introducing pathway information significantly improves the learning of cell-cell graphs, with their resulting consensus graph enhancing the predictive performance of cell type prediction. Different pathway databases provide complementary data, and an increase in the number of pathways can also boost model performance. Extensive testing shows that in various cross-dataset application scenarios, scMCGraph consistently exhibits both accuracy and robustness.

scRSSL: Residual semi-supervised learning with deep generative models to automatically identify cell types

Single-cell sequencing (scRNA-seq) allows researchers to study cellular heterogeneity in individual cells. In single-cell transcriptomics analysis, identifying the cell type of individual cells is a key task. At present, single-cell datasets often face the challenges of high dimensionality, large number of samples, high sparsity and sample imbalance. The traditional methods of cell type recognition have been challenged. The authors propose a deep residual generation model based on semi-supervised learning (scRSSL) to address these challenges. ScRSSL creatively introduces residual networks into semi-supervised generative models. The authors take advantage of its semi-supervised learning to solve the problem of sample imbalance. During the training of the model, the authors use a residual neural network to accomplish the inference of cell types so that local features of single-cell data can be extracted. Because of the semi-supervised learning approach, it can automatically and accurately predict individual cell types in datasets, even with only a small number of cell labels. Experimentally, the authors' method has proven to have better performance compared to other methods.

GDNF family receptor alpha-like (GFRAL) expression is restricted to the caudal brainstem

OBJECTIVE

Growth differentiation factor 15 (GDF15) acts on the receptor dimer of GDNF family receptor alpha-like (GFRAL) and Rearranged during transfection (RET). While Gfral-expressing cells are known to be present in the area postrema and nucleus of the solitary tract (AP/NTS) located in the brainstem, the presence of Gfral-expressing cells in other sites within the central nervous system and peripheral tissues is not been fully addressed. Our objective was to thoroughly investigate whether GFRAL is expressed in peripheral tissues and in brain sites different from the brainstem.

METHODS

From Gfral:eGFP mice we collected tissue from 12 different tissues, including brain, and used single molecule in-situ hybridizations to identify cells within those tissues expressing Gfral. We then contrasted the results with human Gfral-expression by analyzing publicly available single-cell RNA sequencing data.

RESULTS

In mice we found readably detectable Gfral mRNA within the AP/NTS but not within other brain sites. Within peripheral tissues, we failed to detect any Gfral-labelled cells in the vast majority of examined tissues and when present, were extremely rare. Single cell sequencing of human tissues confirmed GFRAL-expressing cells are detectable in some sites outside the AP/NTS in an extremely sparse manner. Importantly, across the utilized methodologies, smFISH, genetic Gfral reporter mice and scRNA-Seq, we failed to detect Gfral-labelled cells with all three.

CONCLUSIONS

Through highly sensitive and selective technologies we show Gfral expression is overwhelmingly restricted to the brainstem and expect that GDF15 and GFRAL-based therapies in development for cancer cachexia will specifically target AP/NTS cells.

pscAdapt: Pre-Trained Domain Adaptation Network Based on Structural Similarity for Cell Type Annotation in Single Cell RNA-seq Data

Cell type annotation refers to the process of categorizing and labeling cells to identify their specific cell types, which is crucial for understanding cell functions and biological processes. Although many methods have been developed for automated cell type annotation, they often encounter challenges such as batch effects due to variations in data distribution across platforms and species, thereby compromising their performance. To address batch effects, in this study, a pre-trained domain adaptation model based on structural similarity, named pscAdapt, is proposed for cell type annotation. Specifically, a pre-trained strategy is employed to initialize model parameters to learn the data distribution of source domain. This strategy is also combined with an adversarial learning strategy to train the domain adaptation network for achieving domain level alignment and reducing domain discrepancy. Furthermore, to better distinguish different types of cells, a structural similarity loss is designed, aiming to shorten distances between cells of the same type and increase distances between cells of different types in feature space, thus achieving cell level alignment and enhancing the discriminability of cell types. Comprehensive experiments were conducted on simulated datasets, cross-platforms datasets and cross-species datasets to validate the effectiveness of pscAdapt, results of which demonstrate that pscAdapt outperforms several popular cell type annotation methods.

Fam3a-mediated prohormone convertase switch in α-cells regulates pancreatic GLP-1 production in an Nr4a2-Foxa2-dependent manner

BACKGROUND

Fam3a has been demonstrated to regulate pancreatic β-cell function and glucose homeostasis. However, the role and mechanism of Fam3a in regulating α-cell function remain unexplored.

METHODS

Glucagon and glucagon-like peptide-1 (GLP-1) levels in pancreas and plasma were measured in global Fam3a knockout (Fam3a-/-) mice. Human islet single-cell RNA sequencing (scRNA-seq) datasets were utilized to analyze gene expression correlations between FAM3A and PCSK1 (encoding PC1/3, which processes proglucagon into GLP-1). Mouse pancreatic α-cell line αTC1.9 cells were transfected with Fam3a siRNA or plasmid for Fam3a knockdown or overexpression to explore the effects of Fam3a on PC1/3 expression and GLP-1 production. The downstream mediator (including Nr4a2) was identified by transcriptomic analysis, and its role was confirmed by Fam3a knockdown or overexpression in αTC1.9 cells. Based on the interacted protein of Nr4a2 and the direct binding to Pcsk1 promoter, the transcription factor Foxa2 was selected for further verification. Nuclear translocation assay and dual-luciferase reporter assay were used to clarify the involvement of Fam3a-Nr4a2-Foxa2 pathway in PC1/3 expression and GLP-1 production. Moreover, α-cell-specific Fam3a knockout (Fam3aα-/-) mice were constructed to evaluate the metabolic variables and hormone levels under normoglycemic, high-fat diet (HFD)-fed and streptozotocin (STZ)-induced diabetic conditions. Exendin 9-39 (Ex9), a GLP-1 receptor antagonist, was used to investigate GLP-1 paracrine effects in Fam3aα-/- mice and in their primary islets.

RESULTS

Compared with wild-type mice, pancreatic and plasma active GLP-1 levels were increased in Fam3a-/- mice. Analysis of human islet scRNA-seq datasets showed a significant negative correction between FAM3A and PCSK1 in α-cells. Fam3a knockdown upregulated PC1/3 expression and GLP-1 production in αTC1.9 cells, while Fam3a overexpression displayed inverse effects. Transcriptomic analysis identified Nr4a2 as a key downstream mediator of Fam3a, and Nr4a2 expression in αTC1.9 cells was downregulated and upregulated by Fam3a knockdown and overexpression, respectively. Nr4a2 silencing increased PC1/3 expression, albeit Nr4a2 did not directly bind to Pcsk1 promoter. Instead, Nr4a2 formed a complex with Foxa2 to facilitate Fam3a-mediated Foxa2 nuclear translocation. Foxa2 negatively regulated PC1/3 expression and GLP-1 production. Besides, Foxa2 inhibited the transcriptional activity of Pcsk1 promoter at specific binding sites 10 and 6, and this inhibition was intensified by Nr4a2 in αTC1.9 cells. Compared with Flox/cre littermates, improved glucose tolerance, increased active GLP-1 level in pancreas and plasma, upregulated plasma insulin level in response to glucose, and decreased plasma glucagon level were observed in Fam3aα-/- mice. Primary islets isolated from Fam3aα-/- mice also showed an increase in active GLP-1 and insulin release. In addition, the insulinotropic effect of intra-islet GLP-1 was blocked by Ex9 in Fam3aα-/- mice and in their primary islets. Similarly, HFD-fed Fam3aα-/- mice also exhibited an improved glucose tolerance. Both HFD-fed and STZ-induced diabetic Fam3aα-/- mice showed an increased pancreatic active GLP-1 level, an elevated plasma insulin level and a reduced plasma glucagon level.

CONCLUSIONS

Fam3a deficiency in α-cells enhances pancreatic GLP-1 production to improve β-cell function via paracrine signaling in an Nr4a2-Foxa2-PC1/3-dependent manner. Our study unveils a novel strategy for reprogramming α-cell proglucagon processing output from glucagon to GLP-1 and deepen the understanding of crosstalk between α-cells and β-cells.

β cell acetate production and release are negligible

BACKGROUND

Studies suggest that short chain fatty acids (SCFAs), which are primarily produced from fermentation of fiber, regulate insulin secretion through free fatty acid receptors 2 and 3 (FFA2 and FFA3). As these are G-protein coupled receptors (GPCRs), they have potential therapeutic value as targets for treating type 2 diabetes (T2D). The exact mechanism by which these receptors regulate insulin secretion and other aspects of pancreatic β cell function is unclear. It has been reported that glucose-dependent release of acetate from pancreatic β cells negatively regulates glucose stimulated insulin secretion. While these data raise the possibility of acetate's potential autocrine action on these receptors, these findings have not been independently confirmed, and multiple concerns exist with this observation, particularly the lack of specificity and precision of the acetate detection methodology used.

METHODS

Using Min6 cells and mouse islets, we assessed acetate and pyruvate production and secretion in response to different glucose concentrations, via liquid chromatography mass spectrometry.

RESULTS

Using Min6 cells and mouse islets, we showed that both intracellular pyruvate and acetate increased with high glucose conditions; however, intracellular acetate level increased only slightly and exclusively in Min6 cells but not in the islets. Further, extracellular acetate levels were not affected by the concentration of glucose in the incubation medium of either Min6 cells or islets.

CONCLUSIONS

Our findings do not substantiate the glucose-dependent release of acetate from pancreatic β cells, and therefore, invalidate the possibility of an autocrine inhibitory effect on glucose stimulated insulin secretion.

Alleviating batch effects in cell type deconvolution with SCCAF-D

Cell type deconvolution methods can impute cell proportions from bulk transcriptomics data, revealing changes in disease progression or organ development. But benchmarking studies often use simulated bulk data from the same source as the reference, which limits its application scenarios. This study examines batch effects in deconvolution and introduces SCCAF-D, a computational workflow that ensures a Pearson Correlation Coefficient above 0.75 across simulated and real bulk data for various tissue types. Applied to non-alcoholic fatty liver disease, SCCAF-D unveils meaningful insights into changes in cell proportions during disease progression.

Pancreatic β cell interleukin-22 receptor subunit alpha 1 deficiency impairs β cell function in type 2 diabetes via cytochrome b5 reductase 3

Impaired β cell function is a hallmark of type 2 diabetes (T2D), but the underlying cellular signaling machineries that regulate β cell function remain unknown. Here, we identify that the interleukin-22 receptor subunit alpha 1 (IL-22RA1), known as a co-receptor for IL-22, is downregulated in human and mouse T2D β cells. Mice with β cell Il22ra1 knockout (Il22ra1βKO) exhibit defective insulin secretion and impaired glucose tolerance after being fed a high-fat diet (HFD) or an HFD/low dose of streptozotocin (STZ). Mechanistically, β cell IL-22RA1 deficiency inhibits cytochrome b5 reductase 3 (CYB5R3) expression via the IL-22RA1/signal transducer and activator of the transcription 3 (STAT3)/c-Jun axis, thereby impairing mitochondrial function and reducing β cell identity. Overexpression of CYB5R3 reinstates mitochondrial function, β cell identity, and insulin secretion in Il22ra1βKO mice. Moreover, the pharmacological activation of CYB5R3 with tetrahydroindenoindole restores insulin secretion in Il22ra1βKO mice, IL-22RA1-knockdown human islets, and Min6 cells. In conclusion, these findings suggest an important role of IL-22RA1 in preserving β cell function in T2D, which offers a potential therapeutic target for treating diabetes.

CORTADO: Hill Climbing Optimization for Cell-Type Specific Marker Gene Discovery

The advent of single-cell RNA sequencing (scRNA-seq) has greatly enhanced our ability to explore cellular heterogeneity with high resolution. Identifying subpopulations of cells and their associated molecular markers is crucial in understanding their distinct roles in tissues. To address the challenges in marker gene selection, we introduce CORTADO, a computational framework based on hill-climbing optimization for the efficient discovery of cell-type-specific markers. CORTADO optimizes three critical properties: differential expression in the clusters of interest, distinctiveness in gene expression profiles to minimize redundancy, and sparseness to ensure a concise and biologically meaningful marker set. Unlike traditional methods that rely on ranking genes by p-values, CORTADO incorporates both differential expression metrics and penalties for overlapping expression profiles, ensuring that each selected marker uniquely represents its cluster while maintaining biological relevance. Its flexibility supports both constrained and unconstrained marker selection, allowing users to specify the number of markers to identify, making it adaptable to diverse analytical needs and scalable to datasets with varying complexities. To validate its performance, we apply CORTADO to several datasets, including the DLPFC 151507 dataset, the Zeisel mouse brain dataset, and a peripheral blood mononuclear cell dataset. Through enrichment analysis and examination of spatial localization-based expression, we demonstrate the robustness of CORTADO in identifying biologically relevant and non-redundant markers in complex datasets. CORTADO provides an efficient and scalable solution for cell-type marker discovery, offering improved sensitivity and specificity compared to existing methods.

CFTR represses a PDX1 axis to govern pancreatic ductal cell fate

Inflammation, acinar atrophy, and ductal hyperplasia drive pancreatic remodeling in newborn cystic fibrosis (CF) ferrets lacking a functional cystic fibrosis conductance regulator (CFTR) channel. These changes are associated with a transient phase of glucose intolerance that involves islet destruction and subsequent regeneration near hyperplastic ducts. The phenotypic changes in CF ductal epithelium and their impact on islet function are unknown. Using bulk RNA sequencing (RNA-seq), single-cell RNA sequencing (scRNA-seq), and assay for transposase-accessible chromatin using sequencing (ATAC-seq) on CF ferret models, we demonstrate that ductal CFTR protein constrains PDX1 expression by maintaining PTEN and GSK3β activation. In the absence of CFTR protein, centroacinar cells adopted a bipotent progenitor-like state associated with enhanced WNT/β-Catenin, transforming growth factor β (TGF-β), and AKT signaling. We show that the level of CFTR protein, not its channel function, regulates PDX1 expression. Thus, this study has discovered a cell-autonomous CFTR-dependent mechanism by which CFTR mutations that produced little to no protein could impact pancreatic exocrine/endocrine remodeling in people with CF.

Genetic inactivation of zinc transporter SLC39A5 improves liver function and hyperglycemia in obesogenic settings

Recent studies have revealed a role for zinc in insulin secretion and glucose homeostasis. Randomized placebo-controlled zinc supplementation trials have demonstrated improved glycemic traits in patients with type II diabetes (T2D). Moreover, rare loss-of-function variants in the zinc efflux transporter SLC30A8 reduce T2D risk. Despite this accumulated evidence, a mechanistic understanding of how zinc influences systemic glucose homeostasis and consequently T2D risk remains unclear. To further explore the relationship between zinc and metabolic traits, we searched the exome database of the Regeneron Genetics Center-Geisinger Health System DiscovEHR cohort for genes that regulate zinc levels and associate with changes in metabolic traits. We then explored our main finding using in vitro and in vivo models. We identified rare loss-of-function (LOF) variants (MAF <1%) in Solute Carrier Family 39, Member 5 (SLC39A5) associated with increased circulating zinc (p=4.9 × 10-4). Trans-ancestry meta-analysis across four studies exhibited a nominal association of SLC39A5 LOF variants with decreased T2D risk. To explore the mechanisms underlying these associations, we generated mice lacking Slc39a5. Slc39a5-/- mice display improved liver function and reduced hyperglycemia when challenged with congenital or diet-induced obesity. These improvements result from elevated hepatic zinc levels and concomitant activation of hepatic AMPK and AKT signaling, in part due to zinc-mediated inhibition of hepatic protein phosphatase activity. Furthermore, under conditions of diet-induced non-alcoholic steatohepatitis (NASH), Slc39a5-/- mice display significantly attenuated fibrosis and inflammation. Taken together, these results suggest SLC39A5 as a potential therapeutic target for non-alcoholic fatty liver disease (NAFLD) due to metabolic derangements including T2D.

Analyzing scRNA-seq data by CCP-assisted UMAP and tSNE

Single-cell RNA sequencing (scRNA-seq) is widely used to reveal heterogeneity in cells, which has given us insights into cell-cell communication, cell differentiation, and differential gene expression. However, analyzing scRNA-seq data is a challenge due to sparsity and the large number of genes involved. Therefore, dimensionality reduction and feature selection are important for removing spurious signals and enhancing downstream analysis. Correlated clustering and projection (CCP) was recently introduced as an effective method for preprocessing scRNA-seq data. CCP utilizes gene-gene correlations to partition the genes and, based on the partition, employs cell-cell interactions to obtain super-genes. Because CCP is a data-domain approach that does not require matrix diagonalization, it can be used in many downstream machine learning tasks. In this work, we utilize CCP as an initialization tool for uniform manifold approximation and projection (UMAP) and t-distributed stochastic neighbor embedding (tSNE). By using 21 publicly available datasets, we have found that CCP significantly improves UMAP and tSNE visualization and dramatically improve their accuracy. More specifically, CCP improves UMAP by 22% in ARI, 14% in NMI and 15% in ECM, and improves tSNE by 11% in ARI, 9% in NMI and 8% in ECM.