Explore the vast range of titles available.

Recent advances in multiplexed single‐cell transcriptomics experiments facilitate the high‐throughput study of drug and genetic perturbations. However, an exhaustive exploration of the combinatorial perturbation space is experimentally unfeasible. Therefore, computational methods are needed to predict, interpret, and prioritize perturbations. Here, we present the compositional perturbation autoencoder (CPA), which combines the interpretability of linear models with the flexibility of deep‐learning approaches for single‐cell response modeling. CPA learns to in silico predict transcriptional perturbation response at the single‐cell level for unseen dosages, cell types, time points, and species. Using newly generated single‐cell drug combination data, we validate that CPA can predict unseen drug combinations while outperforming baseline models. Additionally, the architecture's modularity enables incorporating the chemical representation of the drugs, allowing the prediction of cellular response to completely unseen drugs. Furthermore, CPA is also applicable to genetic combinatorial screens. We demonstrate this by imputing in silico 5,329 missing combinations (97.6% of all possibilities) in a single‐cell Perturb‐seq experiment with diverse genetic interactions. We envision CPA will facilitate efficient experimental design and hypothesis generation by enabling in silico response prediction at the single‐cell level and thus accelerate therapeutic applications using single‐cell technologies. Synopsis The compositional perturbation autoencoder (CPA) is a deep learning model for predicting the transcriptomic responses of single cells to single or combinatorial treatments from drugs and genetic manipulations. CPA can be trained on highly multiplexed, single‐cell experiments with thousands of conditions to predict unmeasured phenotypes (e.g., specific dose responses). It can generalize to predict responses to small molecules never seen in the training by adding priors on chemical space. Validations using a newly generated combinatorial drug perturbation dataset demonstrate the accuracy of CPA in predicting unseen drug combinations. CPA is also applicable to genetic combinatorial screens, as shown by imputing in silico 5,329 missing combinations in a single‐cell perturb‐seq experiment with diverse genetic interactions. Graphical Abstract The compositional perturbation autoencoder (CPA) is a deep learning model for predicting the transcriptomic responses of single cells to single or combinatorial treatments from drugs and genetic manipulations.

Tissue dissociation, a crucial step in single‐cell sample preparation, can alter the transcriptional state of a sample through the intrinsic cellular stress response. Here we demonstrate a general approach for measuring transcriptional response during sample preparation. In our method, transcripts made during dissociation are labeled for later identification upon sequencing. We found general as well as cell‐type‐specific dissociation response programs in zebrafish larvae, and we observed sample‐to‐sample variation in the dissociation response of mouse cardiomyocytes despite well‐controlled experimental conditions. Finally, we showed that dissociation of the mouse hippocampus can lead to the artificial activation of microglia. In summary, our approach facilitates experimental optimization of dissociation procedures as well as computational removal of transcriptional perturbation response. Synopsis A new approach shows that RNA labelling can be used to measure the cellular response to tissue dissociation, a major confounding factor in single‐cell transcriptomics. The dissociation response is partially cell‐type specific. Single‐cell RNA labelling allows measuring the cellular dissociation response, a major confounding factor in single‐cell transcriptomics. The dissociation response is comprised of a core signature that is shared across tissue types and replicates as well as sample‐ and cell‐type‐specific programs. The dissociation of the mouse hippocampus can lead to the activation of microglia. Graphical Abstract A new approach shows that RNA labelling can be used to measure the cellular response to tissue dissociation, a major confounding factor in single‐cell transcriptomics. The dissociation response is partially cell‐type specific.

Human intestinal epithelial cells form a primary barrier protecting us from pathogens, yet only limited knowledge is available about individual contribution of each cell type to mounting an immune response against infection. Here, we developed a framework combining single‐cell RNA‐Seq and highly multiplex RNA FISH and applied it to human intestinal organoids infected with human astrovirus, a model human enteric virus. We found that interferon controls the infection and that astrovirus infects all major cell types and lineages and induces expression of the cell proliferation marker MKI67. Intriguingly, each intestinal epithelial cell lineage exhibits a unique basal expression of interferon‐stimulated genes and, upon astrovirus infection, undergoes an antiviral transcriptional reprogramming by upregulating distinct sets of interferon‐stimulated genes. These findings suggest that in the human intestinal epithelium, each cell lineage plays a unique role in resolving virus infection. Our framework is applicable to other organoids and viruses, opening new avenues to unravel roles of individual cell types in viral pathogenesis. SYNOPSIS Single‐cell sequencing and multiplex single‐molecule RNA FISH analyses of human astrovirus 1 (HAstV1)‐infected human intestinal organoids characterize viral tropism and unravel the cell lineage‐specific immune response to viral infection. A single‐cell RNA‐Seq reference dataset of human ileum biopsies is established. An integrative framework to investigate cell‐type‐specific viral pathogenesis in a tissue‐like environment is developed. HAstV1 infects all lineages from the human intestinal epithelium, causing an interferon‐mediated immune response. HAstV1 evokes a cell lineage‐specific intrinsic immune response. Each intestinal cell lineage has a different steady expression of interferon‐stimulated genes (ISGs). Graphical Abstract Single‐cell sequencing and multiplex single‐molecule RNA FISH analyses of human astrovirus 1 (HAstV1)‐infected human intestinal organoids characterize viral tropism and unravel the cell lineage‐specific immune response to viral infection.

Cancer‐associated fibroblasts (CAFs) are sought after as potential therapeutic targets due to their pro‐ and antitumorigenic functions, which are attributed to specializations in CAF subtypes. A precise targeting of specific subtypes would be required to design therapies that effectively modulate CAF phenotypes, necessitating translatable model systems to support target discovery efforts. However, not only is our knowledge of CAF heterogeneity in solid tumors lacking, particularly in pancreatic tumors, but the translatability of CAF models has also not been rigorously evaluated. Here, we develop a coculturing model with primary CAFs and immortalized tumor cell lines that can reliably represent CAF phenotypes observed in tumors, with correlations to immuno‐resistant and immunomodulatory phenotypes. Using single‐cell transcriptomics, we characterize the CAF subtype heterogeneity in the in vitro CAF cell lines isolated from pancreatic cancer patients and investigate the impact of perturbing potential stromal genes on different CAF subtypes. We also infer the continuum of state changes underlying the interconvertibility of CAF subtypes. Finally, we use immortalized CAF cell lines to perform single‐cell CRISPR perturbations of stromal targets, revealing the subtype‐specific effects of perturbations and the impact of model‐type selection on the translatability of insights. Development of therapies targeting cancer‐associated fibroblasts (CAFs) necessitates preclinical model systems that faithfully represent CAF–tumor biology. We established an in vitro coculture system of patient‐derived pancreatic CAFs and tumor cell lines and demonstrated its recapitulation of primary CAF–tumor biology with single‐cell transcriptomics and phenotypic validation. With Perturb‐seq, we showed that the model enabled studies on subtype‐specific effects of genetic perturbations.

Summary Ligand‐receptor pairs play important roles in cell–cell communication for multicellular organisms in response to environmental cues. Recently, the emergence of single‐cell RNA‐sequencing (scRNA‐seq) provides unprecedented opportunities to investigate cellular communication based on ligand‐receptor expression. However, so far, no reliable ligand‐receptor interaction database is available for plant species. In this study, we developed PlantPhoneDB (https://jasonxu.shinyapps.io/PlantPhoneDB/), a pan‐plant database comprising a large number of high‐confidence ligand‐receptor pairs manually curated from seven resources. Also, we developed a PlantPhoneDB R package, which not only provided optional four scoring approaches that calculate interaction scores of ligand‐receptor pairs between cell types but also provided visualization functions to present analysis results. At the PlantPhoneDB web interface, the processed datasets and results can be searched, browsed, and downloaded. To uncover novel cell–cell communication events in plants, we applied the PlantPhoneDB R package on GSE121619 dataset to infer significant cell–cell interactions of heat‐shocked root cells in Arabidopsis thaliana. As a result, the PlantPhoneDB predicted the actively communicating AT1G28290‐AT2G14890 ligand‐receptor pair in atrichoblast–cortex cell pair in Arabidopsis thaliana. Importantly, the downstream target genes of this ligand‐receptor pair were significantly enriched in the ribosome pathway, which facilitated plants adapting to environmental changes. In conclusion, PlantPhoneDB provided researchers with integrated resources to infer cell–cell communication from scRNA‐seq datasets.

The oral mucosa exhibits superior healing and minimal scarring. Although mouse models are widely used to study wound healing and various diseases, their translational relevance remains unclear. Here, we performed a comparative single‐cell transcriptomic analysis of human and mouse oral mucosa to identify both shared and species‐specific mechanisms. A total of 34,969 cells from human and mouse datasets were integrated using Harmony for batch effect correction, allowing us to establish a unified oral mucosa transcriptome atlas. Fibroblasts emerged as the prominent cell population in both species, displaying conserved gene expression profiles and cell communication networks, underscoring their central role in tissue homeostasis. Key pathways involved in extracellular matrix remodelling and wound healing were highly conserved, supporting the utility of mouse models for studying fibroblast‐mediated tissue regeneration. These findings suggest that mouse models can effectively replicate human fibroblast biology, offering valuable insights for developing translational therapies that target fibroblast activity and regulatory gene networks to enhance wound healing and tissue regeneration. Additionally, we identified species‐specific cell populations, including human‐specific capillary endothelial cells and melanocytes, as well as mouse‐specific salivary gland epithelial cells. Their distinct cellular composition and functional differences suggest that these subpopulations may not be directly translatable from mouse models to human contexts. Overall, our study highlights the evolutionary conservation of fibroblasts while identifying species‐specific differences that warrant consideration in translational research. These findings provide a valuable resource for researchers using mouse models to study oral mucosa‐related diseases, facilitating the translation of preclinical discoveries into clinical applications.

Activating autologous stem cells after the implantation of biomaterials is an important process to initiate bone regeneration. Although several studies have demonstrated the mechanism of biomaterial‐mediated bone regeneration, a comprehensive single‐cell level transcriptomic map revealing the influence of biomaterials on regulating the temporal and spatial expression patterns of mesenchymal stem cells (MSCs) is still lacking. Herein, the osteoimmune microenvironment is depicted around the classical collagen/nanohydroxyapatite‐based bone repair materials via combining analysis of single‐cell RNA sequencing and spatial transcriptomics. A group of functional MSCs with high expression of matrix Gla protein (Mgp) is identified, which may serve as a pioneer subpopulation involved in bone repair. Remarkably, these Mgp high‐expressing MSCs (MgphiMSCs) exhibit efficient osteogenic differentiation potential and orchestrate the osteoimmune microenvironment around implanted biomaterials, rewiring the polarization and osteoclastic differentiation of macrophages through the Mdk/Lrp1 ligand–receptor pair. The inhibition of Mdk/Lrp1 activates the pro‐inflammatory programs of macrophages and osteoclastogenesis. Meanwhile, multiple immune‐cell subsets also exhibit close crosstalk between MgphiMSCs via the secreted phosphoprotein 1 (SPP1) signaling pathway. These cellular profiles and interactions characterized in this study can broaden the understanding of the functional MSC subpopulations at the early stage of biomaterial‐mediated bone regeneration and provide the basis for materials‐designed strategies that target osteoimmune modulation. This study reveals how collagen/nanohydroxyapatite‐based bone repair materials regulate the osteoimmune microenvironment via combining single‐cell RNA sequencing and spatial transcriptomics analyses. The Mgp high‐expressing MSCs with efficient osteogenic differentiation potential are found to be aggregated around implanted materials at the early stage of bone regeneration, regulating the polarization and osteoclastic differentiation of macrophages through the Mdk/Lrp1 ligand–receptor pair.

Infection by the fungal pathogen Magnaporthe oryzae elicits dynamic responses in rice. Utilizing an integrated approach of single‐cell and spatial transcriptomics, a 3D response is uncovered within rice leaf cells to M. oryzae infection. A comprehensive rice leaf atlas is constructed from 236 708 single‐cell transcriptomes, revealing heightened expression of immune receptors, namely Pattern Recognition Receptors (PRRs) and Nucleotide‐binding site and leucine‐rich repeat (NLRs) proteins, within vascular tissues. Diterpene phytoalexins biosynthesis genes are dramatically upregulated in procambium cells, leading to an accumulation of these phytoalexins within vascular bundles. Consistent with these findings, microscopic observations confirmed that M. oryzae is prone to target leaf veins for invasion, yet is unable to colonize further within vascular tissues. Following fungal infection, basal defenses are extensively activated in rice cells, as inferred from trajectory analyses. The spatial transcriptomics reveals that rice leaf tissues toward leaf tips display stronger immunity. Characterization of the polarity gene OsHKT9 suggests that potassium transport plays a critical role in resisting M. oryzae infection by expression along the longitudinal axis, where the immunity is stronger toward leaf tip. This work uncovers that there is a cell‐specific and multi‐dimensional (local and longitudinal) immune response to a fungal pathogen infection. By employing a combination of single cell and spatial transcriptomic sequencing, this study presents a stereoscopic response of rice leaf to Magnaporthe oryzae infection. The vascular tissues mount defenses by producing phytoalexins. The immune strength is stronger toward the rice leaf tip than that of the leaf base.

Summary Southern corn rust (SCR), caused by Puccinia polysora Underw (P. polysora), is a catastrophic disease affecting maize, leading to significant global yield losses. The disease manifests primarily as pustules on the upper surface of corn leaves, obscuring our understanding of its cellular heterogeneity, the maize's response to its infection and the underlying gene expression regulatory mechanisms. In this study, we dissected the heterogeneity of maize's response to P. polysora infection using single‐cell RNA sequencing. We delineated cell‐type‐specific gene expression alterations in six leaf cell types, creating the inaugural single‐cell atlas of a maize leaf under fungal assault. Crucially, by reconstructing cellular trajectories in susceptible line N110 and resistant line R99 during infection, we identified diverse regulatory programs that fortify R99's resistance across different leaf cell types. This research uncovers an immune‐like state in R99 leaves, characterized by the expression of various fungi‐induced genes in the absence of fungal infection, particularly in guard and epidermal cells. Our findings also highlight the role of the fungi‐induced glycoside hydrolase family 18 chitinase 7 protein (ZmChit7) in conferring resistance to P. polysora. Collectively, our results shed light on the mechanisms of maize resistance to fungal pathogens through comparative single‐cell transcriptomics, offering a valuable resource for pinpointing novel genes that bolster resistance to P. polysora.