StDiff: a diffusion model for imputing spatial transcriptomics through single-cell transcriptomics
Ontology highlight
ABSTRACT:
SUBMITTER:
PROVIDER: S-EPMC11021815 | biostudies-literature | 2024 Apr
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Image-based spatial transcriptomic sequencing technologies have enabled the measurement of gene expression at single-cell resolution, but with a limited number of genes. Current computational approaches attempt to overcome these limitations by imputing missing genes, but face challenges regarding prediction accuracy and identification of cell populations due to the neglect of gene-gene relationships. In this context, we present stImpute, a method to impute spatial transcriptomics according to reference scRNA-seq data based on the gene network constructed from the protein language model ESM-2. Specifically, stImpute employs an autoencoder to create gene expression embeddings for both spatial transcriptomics and scRNA-seq data, which are used to identify the nearest neighboring cells between scRNA-seq and spatial transcriptomics datasets. According to the neighbored cells, the gene expressions of spatial transcriptomics cells are imputed through a graph neural network, where nodes are genes, and edges are based on cosine similarity between the ESM-2 embeddings of the gene-encoding proteins. The gene prediction uncertainty is further measured through a deep learning model. stImpute was shown to consistently outperform state-of-the-art methods across multiple datasets concerning imputation and clustering. stImpute also demonstrates robustness in producing consistent results that are insensitive to model parameters.
| S-EPMC11455941 | biostudies-literature
Project description:Spatial patterns of gene expression manifest at scales ranging from local (e.g., cell-cell interactions) to global (e.g., body axis patterning). However, current spatial transcriptomics methods either average local contexts or are restricted to limited fields of view. Here, we introduce sci-Space, which retains single-cell resolution while resolving spatial heterogeneity at larger scales. Applying sci-Space to developing mouse embryos, we captured approximate spatial coordinates and whole transcriptomes of about 120,000 nuclei. We identify thousands of genes exhibiting anatomically patterned expression, leverage spatial information to annotate cellular subtypes, show that cell types vary substantially in their extent of spatial patterning, and reveal correlations between pseudotime and the migratory patterns of differentiating neurons. Looking forward, we anticipate that sci-Space will facilitate the construction of spatially resolved single-cell atlases of mammalian development.
| S-EPMC9118175 | biostudies-literature
Project description:The intricate development and functionality of the mammalian heart are influenced by the heterogeneous nature of cardiomyocytes (CMs). In this study, single-cell and spatial transcriptomics were utilized to analyze cells from neonatal mouse hearts, resulting in a comprehensive atlas delineating the transcriptional profiles of distinct CM subsets. A continuum of maturation states was elucidated, emphasizing a progressive developmental trajectory rather than discrete stages. This approach enabled the mapping of these states across various cardiac regions, illuminating the spatial organization of CM development and the influence of the cellular microenvironment. Notably, a subset of transitional CMs was identified, characterized by a transcriptional signature marking a pivotal maturation phase, presenting a promising target for therapeutic strategies aimed at enhancing cardiac regeneration. This atlas not only elucidates fundamental aspects of cardiac development but also serves as a valuable resource for advancing research into cardiac physiology and pathology, with significant implications for regenerative medicine.
| S-EPMC11732507 | biostudies-literature
Project description:Single-cell technologies allow measuring chromatin accessibility and gene expression in each cell, but jointly utilizing both layers to map bona fide gene regulatory networks and enhancers remains challenging. Here, we generate independent single-cell RNA-seq and single-cell ATAC-seq atlases of the Drosophila eye-antennal disc and spatially integrate the data into a virtual latent space that mimics the organization of the 2D tissue using ScoMAP (Single-Cell Omics Mapping into spatial Axes using Pseudotime ordering). To validate spatially predicted enhancers, we use a large collection of enhancer-reporter lines and identify ~ 85% of enhancers in which chromatin accessibility and enhancer activity are coupled. Next, we infer enhancer-to-gene relationships in the virtual space, finding that genes are mostly regulated by multiple, often redundant, enhancers. Exploiting cell type-specific enhancers, we deconvolute cell type-specific effects of bulk-derived chromatin accessibility QTLs. Finally, we discover that Prospero drives neuronal differentiation through the binding of a GGG motif. In summary, we provide a comprehensive spatial characterization of gene regulation in a 2D tissue.
| S-EPMC7237818 | biostudies-literature
Project description:BackgroundIntratumour heterogeneity is a hallmark of most solid tumours, including breast cancers. We applied spatial transcriptomics and single-cell RNA-sequencing on patient-derived xenografts (PDXs) to profile spatially resolved cell populations within oestrogen receptor-positive (ER+ ) breast cancer and to elucidate their importance in oestrogen-dependent tumour growth.MethodsTwo PDXs of 'ER-high' breast cancers with opposite oestrogen-mediated growth responses were investigated: oestrogen-suppressed GS3 (80-100% ER) and oestrogen-dependent SC31 (40-90% ER) models. The observation was validated via single-cell analyses on an 'ER-low' PDX, GS1 (5% ER). The results from our spatial and single-cell analyses were further supported by a public ER+ breast cancer single-cell dataset and protein-based dual immunohistochemistry (IHC) of SC31 examining important luminal cancer markers (i.e., ER, progesterone receptor and Ki67). The translational implication of our findings was assessed by clinical outcome analyses on publicly available cohorts.ResultsOur space-gene-function study revealed four spatially distinct compartments within ER+ breast cancers. These compartments showed functional diversity (oestrogen-responsive, proliferative, hypoxia-induced and inflammation-related). The 'proliferative' population, rather than the 'oestrogen-responsive' compartment, was crucial for oestrogen-dependent tumour growth, leading to the acquisition of luminal B-like features. The cells expressing typical oestrogen-responsive genes like PGR were not directly linked to oestrogen-dependent proliferation. Dual IHC analyses demonstrated the distinct contribution of the Ki67+ proliferative cells toward oestrogen-mediated growth and their response to a CDK4/6 inhibitor. The gene signatures derived from the proliferative, hypoxia-induced and inflammation-related compartments were significantly correlated with worse clinical outcomes, while patients with the oestrogen-responsive signature showed better prognoses, suggesting that this compartment would not be directly associated with oestrogen-dependent tumour progression.ConclusionsOur study identified the gene signature in our 'proliferative' compartment as an important determinant of luminal cancer subtypes. This 'proliferative' cell population is a causative feature of luminal B breast cancer, contributing toward its aggressive behaviours.
| S-EPMC10823285 | biostudies-literature
Project description:Deep neural networks have been widely applied for missing data imputation. However, most existing studies have been focused on imputing continuous data, while discrete data imputation is under-explored. Discrete data is common in real world, especially in research areas of bioinformatics, genetics, and biochemistry. In particular, large amounts of recent genomic data are discrete count data generated from single-cell RNA sequencing (scRNA-seq) technology. Most scRNA-seq studies produce a discrete matrix with prevailing 'false' zero count observations (missing values). To make downstream analyses more effective, imputation, which recovers the missing values, is often conducted as the first step in pre-processing scRNA-seq data. In this paper, we propose a novel Zero-Inflated Negative Binomial (ZINB) model-based autoencoder for imputing discrete scRNA-seq data. The novelties of our method are twofold. First, in addition to optimizing the ZINB likelihood, we propose to explicitly model the dropout events that cause missing values by using the Gumbel-Softmax distribution. Second, the zero-inflated reconstruction is further optimized with respect to the raw count matrix. Extensive experiments on simulation datasets demonstrate that the zero-inflated reconstruction significantly improves imputation accuracy. Real data experiments show that the proposed imputation can enhance separating different cell types and improve the accuracy of differential expression analysis.
| S-EPMC8592282 | biostudies-literature
Project description:High-grade serous ovarian carcinoma (HGSOC) poses a formidable clinical challenge due to multidrug resistance (MDR) caused by tumor heterogeneity. To elucidate the intricate mechanisms underlying HGSOC heterogeneity, we conducted a comprehensive analysis of five single-cell transcriptomes and eight spatial transcriptomes derived from eight HGSOC patients. This study provides a comprehensive view of tumor heterogeneity across the spectrum of gene expression, copy number variation (CNV), and single-cell profiles. Our CNV analysis revealed intratumor heterogeneity by identifying distinct tumor clones, illuminating their evolutionary trajectories and spatial relationships. We further explored the homogeneity and heterogeneity of CNV across tumors to pinpoint the origin of heterogeneity. At the cellular level, single-cell RNA sequencing (scRNA seq) analysis identified three meta-programs that delineate the functional profile of tumor cells. The communication networks between tumor cell clusters exhibited unique patterns associated with the meta-programs governing these clusters. Notably, the ligand-receptor pair MDK - NCL emerged as a highly enriched interaction in tumor cell communication. To probe the functional significance of this interaction, we induced NCL overexpression in the SOVK3 cell line and observed enhanced tumor cell proliferation. These findings indicate that the MDK - NCL interaction plays a crucial role in promoting HGSOC tumor growth and may represent a promising therapeutic target. In conclusion, this study comprehensively unravels the multifaceted nature of HGSOC heterogeneity, providing potential therapeutic strategies for this challenging malignancy.
| S-EPMC11878925 | biostudies-literature
Project description:BackgroundCardiac myxoma (CM) is the most common (58%-80%) type of primary cardiac tumours. Currently, there is a need to develop medical therapies, especially for patients not physically suitable for surgeries. However, the mechanisms that shape the tumour microenvironment (TME) in CM remain largely unknown, which impedes the development of targeted therapies. Here, we aimed to dissect the TME in CM at single-cell and spatial resolution.MethodsWe performed single-cell transcriptomic sequencing and Visium CytAssist spatial transcriptomic (ST) assays on tumour samples from patients with CM. A comprehensive analysis was performed, including unsupervised clustering, RNA velocity, clonal substructure inference of tumour cells and cell-cell communication.ResultsUnsupervised clustering of 34 759 cells identified 12 clusters, which were assigned to endothelial cells (ECs), mesenchymal stroma cells (MSCs), and tumour-infiltrating immune cells. Myxoma tumour cells were found to encompass two closely related phenotypic states, namely, EC-like tumour cells (ETCs) and MSC-like tumour cells (MTCs). According to RNA velocity, our findings suggest that ETCs may be directly differentiated from MTCs. The immune microenvironment of CM was found to contain multiple factors that promote immune suppression and evasion, underscoring the potential of using immunotherapies as a treatment option. Hyperactive signals sent primarily by tumour cells were identified, such as MDK, HGF, chemerin, and GDF15 signalling. Finally, the ST assay uncovered spatial features of the subclusters, proximal cell-cell communication, and clonal evolution of myxoma tumour cells.ConclusionsOur study presents the first comprehensive characterisation of the TME in CM at both single-cell and spatial resolution. Our study provides novel insight into the differentiation of myxoma tumour cells and advance our understanding of the TME in CM. Given the rarity of cardiac tumours, our study provides invaluable datasets and promotes the development of medical therapies for CM.
| S-EPMC10844892 | biostudies-literature
Project description:In recent years, the explosive growth of spatial technologies has enabled the characterization of spatial heterogeneity of tissue architectures. Compared to traditional sequencing, spatial transcriptomics reserves the spatial information of each captured location and provides novel insights into diverse spatially related biological contexts. Even though two spatial transcriptomics databases exist, they provide limited analytical information. Information such as spatial heterogeneity of genes and cells, cell-cell communication activities in space, and the cell type compositions in the microenvironment are critical clues to unveil the mechanism of tumorigenesis and embryo differentiation. Therefore, we constructed a new spatial transcriptomics database, named SPASCER (https://ccsm.uth.edu/SPASCER), designed to help understand the heterogeneity of tissue organizations, region-specific microenvironment, and intercellular interactions across tissue architectures at multiple levels. SPASCER contains datasets from 43 studies, including 1082 sub-datasets from 16 organ types across four species. scRNA-seq was integrated to deconvolve/map spatial transcriptomics, and processed with spatial cell-cell interaction, gene pattern and pathway enrichment analysis. Cell-cell interactions and gene regulation network of scRNA-seq from matched spatial transcriptomics were performed as well. The application of SPASCER will provide new insights into tissue architecture and a solid foundation for the mechanistic understanding of many biological processes in healthy and diseased tissues.
| S-EPMC9825565 | biostudies-literature
Project description:BackgroundThe development of single-cell technologies yields large datasets of information as diverse and multimodal as transcriptomes, immunophenotypes, and spatial position from tissue sections in the so-called 'spatial transcriptomics'. Currently however, user-friendly, powerful, and free algorithmic tools for straightforward analysis of spatial transcriptomic datasets are scarce.ResultsHere, we introduce Single-Cell Spatial Explorer, an open-source software for multimodal exploration of spatial transcriptomics, examplified with 9 human and murine tissues datasets from 4 different technologies.ConclusionsSingle-Cell Spatial Explorer is a very powerful, versatile, and interoperable tool for spatial transcriptomics analysis.
| S-EPMC9881287 | biostudies-literature