BGI Research

Spatial transcriptomics technologies generate gene expression profiles with spatial context, requiring spatially informed analysis tools for three key tasks, spatial clustering, multisample integration, and cell-type deconvolution. We present GraphST, a graph self-supervised contrastive learning method that fully exploits spatial transcriptomics data to outperform existing methods. It combines graph neural networks with self-supervised contrastive learning to learn informative and discriminative spot representations by minimizing the embedding distance between spatially adjacent spots and vice versa. We demonstrated GraphST on multiple tissue types and technology platforms. GraphST achieved 10% higher clustering accuracy and better delineated fine-grained tissue structures in brain and embryo tissues. GraphST is also the only method that can jointly analyze multiple tissue slices in vertical or horizontal integration while correcting batch effects. Lastly, GraphST demonstrated superior cell-type deconvolution to capture spatial niches like lymph node germinal centers and exhausted tumor infiltrating T cells in breast tumor tissue.

Optimal integration of transcriptomics data and associated spatial information is essential towards fully exploiting spatial transcriptomics to dissect tissue heterogeneity and map out inter-cellular communications. We present SEDR, which uses a deep autoencoder coupled with a masked self-supervised learning mechanism to construct a low-dimensional latent representation of gene expression, which is then simultaneously embedded with the corresponding spatial information through a variational graph autoencoder. SEDR achieved higher clustering performance on manually annotated 10 × Visium datasets and better scalability on high-resolution spatial transcriptomics datasets than existing methods. Additionally, we show SEDR's ability to impute and denoise gene expression (URL: https://github.com/JinmiaoChenLab/SEDR/ ).

Abstract Despite tremendous efforts in the past decades, relationships among main avian lineages remain heavily debated without a clear resolution. Discrepancies have been attributed to diversity of species sampled, phylogenetic method and the choice of genomic regions 1–3 . Here we address these issues by analysing the genomes of 363 bird species 4 (218 taxonomic families, 92% of total). Using intergenic regions and coalescent methods, we present a well-supported tree but also a marked degree of discordance. The tree confirms that Neoaves experienced rapid radiation at or near the Cretaceous–Palaeogene boundary. Sufficient loci rather than extensive taxon sampling were more effective in resolving difficult nodes. Remaining recalcitrant nodes involve species that are a challenge to model due to either extreme DNA composition, variable substitution rates, incomplete lineage sorting or complex evolutionary events such as ancient hybridization. Assessment of the effects of different genomic partitions showed high heterogeneity across the genome. We discovered sharp increases in effective population size, substitution rates and relative brain size following the Cretaceous–Palaeogene extinction event, supporting the hypothesis that emerging ecological opportunities catalysed the diversification of modern birds. The resulting phylogenetic estimate offers fresh insights into the rapid radiation of modern birds and provides a taxon-rich backbone tree for future comparative studies.

Elucidating the cellular organization of the cerebral cortex is critical for understanding brain structure and function. Using large-scale single-nucleus RNA sequencing and spatial transcriptomic analysis of 143 macaque cortical regions, we obtained a comprehensive atlas of 264 transcriptome-defined cortical cell types and mapped their spatial distribution across the entire cortex. We characterized the cortical layer and region preferences of glutamatergic, GABAergic, and non-neuronal cell types, as well as regional differences in cell-type composition and neighborhood complexity. Notably, we discovered a relationship between the regional distribution of various cell types and the region's hierarchical level in the visual and somatosensory systems. Cross-species comparison of transcriptomic data from human, macaque, and mouse cortices further revealed primate-specific cell types that are enriched in layer 4, with their marker genes expressed in a region-dependent manner. Our data provide a cellular and molecular basis for understanding the evolution, development, aging, and pathogenesis of the primate brain.

The past two decades has witnessed a remarkable increase in the number of microbial genomes retrieved from marine systems1,2. However, it has remained challenging to translate this marine genomic diversity into biotechnological and biomedical applications3,4. Here we recovered 43,191 bacterial and archaeal genomes from publicly available marine metagenomes, encompassing a wide range of diversity with 138 distinct phyla, redefining the upper limit of marine bacterial genome size and revealing complex trade-offs between the occurrence of CRISPR–Cas systems and antibiotic resistance genes. In silico bioprospecting of these marine genomes led to the discovery of a novel CRISPR–Cas9 system, ten antimicrobial peptides, and three enzymes that degrade polyethylene terephthalate. In vitro experiments confirmed their effectiveness and efficacy. This work provides evidence that global-scale sequencing initiatives advance our understanding of how microbial diversity has evolved in the oceans and is maintained, and demonstrates how such initiatives can be sustainably exploited to advance biotechnology and biomedicine. Analysis of 43,191 genomes obtained from publicly available marine bacterial and archaeal metagenome data provides insights into marine bacterial evolution, CRISPR–Cas defence and antibiotic resistance genes, and demonstrates the potential of marine metagenomes for biotechnological applications.

Abstract Muscle atrophy and functional decline (sarcopenia) are common manifestations of frailty and are critical contributors to morbidity and mortality in older people 1 . Deciphering the molecular mechanisms underlying sarcopenia has major implications for understanding human ageing 2 . Yet, progress has been slow, partly due to the difficulties of characterizing skeletal muscle niche heterogeneity (whereby myofibres are the most abundant) and obtaining well-characterized human samples 3,4 . Here we generate a single-cell/single-nucleus transcriptomic and chromatin accessibility map of human limb skeletal muscles encompassing over 387,000 cells/nuclei from individuals aged 15 to 99 years with distinct fitness and frailty levels. We describe how cell populations change during ageing, including the emergence of new populations in older people, and the cell-specific and multicellular network features (at the transcriptomic and epigenetic levels) associated with these changes. On the basis of cross-comparison with genetic data, we also identify key elements of chromatin architecture that mark susceptibility to sarcopenia. Our study provides a basis for identifying targets in the skeletal muscle that are amenable to medical, pharmacological and lifestyle interventions in late life.

Recent technological developments in spatial transcriptomics allow researchers to measure gene expression of cells and their spatial locations at the single-cell level, generating detailed biological insight into biological processes. A comprehensive database could facilitate the sharing of spatial transcriptomic data and streamline the data acquisition process for researchers. Here, we present the Spatial TranscriptOmics DataBase (STOmicsDB), a database that serves as a one-stop hub for spatial transcriptomics. STOmicsDB integrates 218 manually curated datasets representing 17 species. We annotated cell types, identified spatial regions and genes, and performed cell-cell interaction analysis for these datasets. STOmicsDB features a user-friendly interface for the rapid visualization of millions of cells. To further facilitate the reusability and interoperability of spatial transcriptomic data, we developed standards for spatial transcriptomic data archiving and constructed a spatial transcriptomic data archiving system. Additionally, we offer a distinctive capability of customizing dedicated sub-databases in STOmicsDB for researchers, assisting them in visualizing their spatial transcriptomic analyses. We believe that STOmicsDB could contribute to research insights in the spatial transcriptomics field, including data archiving, sharing, visualization and analysis. STOmicsDB is freely accessible at https://db.cngb.org/stomics/.

Advances in spatial omics technologies now allow multiple types of data to be acquired from the same tissue slice. To realize the full potential of such data, we need spatially informed methods for data integration. Here, we introduce SpatialGlue, a graph neural network model with a dual-attention mechanism that deciphers spatial domains by intra-omics integration of spatial location and omics measurement followed by cross-omics integration. We demonstrated SpatialGlue on data acquired from different tissue types using different technologies, including spatial epigenome-transcriptome and transcriptome-proteome modalities. Compared to other methods, SpatialGlue captured more anatomical details and more accurately resolved spatial domains such as the cortex layers of the brain. Our method also identified cell types like spleen macrophage subsets located at three different zones that were not available in the original data annotations. SpatialGlue scales well with data size and can be used to integrate three modalities. Our spatial multi-omics analysis tool combines the information from complementary omics modalities to obtain a holistic view of cellular and tissue properties.

Abstract Background Cancer‐associated fibroblasts (CAFs) are potential targets for cancer therapy. Due to the heterogeneity of CAFs, the influence of CAF subpopulations on the progression of lung cancer is still unclear, which impedes the translational advances in targeting CAFs. Methods We performed single‐cell RNA sequencing (scRNA‐seq) on tumour, paired tumour‐adjacent, and normal samples from 16 non‐small cell lung cancer (NSCLC) patients. CAF subpopulations were analyzed after integration with published NSCLC scRNA‐seq data. SpaTial enhanced resolution omics‐sequencing (Stereo‐seq) was applied in tumour and tumour‐adjacent samples from seven NSCLC patients to map the architecture of major cell populations in tumour microenvironment (TME). Immunohistochemistry (IHC) and multiplexed IHC (mIHC) were used to validate marker gene expression and the association of CAFs with immune infiltration in TME. Results A subcluster of myofibroblastic CAFs, POSTN + CAFs, were significantly enriched in advanced tumours and presented gene expression signatures related to extracellular matrix remodeling, tumour invasion pathways and immune suppression. Stereo‐seq and mIHC demonstrated that POSTN + CAFs were in close localization with SPP1 + macrophages and were associated with the exhausted phenotype and lower infiltration of T cells. POSTN expression or the abundance of POSTN + CAFs were associated with poor prognosis of NSCLC. Conclusions Our study identified a myofibroblastic CAF subpopulation, POSTN + CAFs, which might associate with SPP1 + macrophages to promote the formation of desmoplastic architecture and participate in immune suppression. Furthermore, we showed that POSTN + CAFs associated with cancer progression and poor clinical outcomes and may provide new insights on the treatment of NSCLC.

Huanglongbing (HLB) is a devastating citrus disease. In this work, we report an HLB resistance regulatory circuit in Citrus composed of an E3 ubiquitin ligase, PUB21, and its substrate, the MYC2 transcription factor, which regulates jasmonate-mediated defense responses. A helitron insertion in the PUB21 promoter introduced multiple MYC2-binding cis-elements to create a regulatory circuit linking the PUB21 activity with MYC2 degradation. Ectopic expression of a natural dominant-negative PUB21 paralog discovered in distant Citrus relatives stabilized MYC2 and conferred resistance to HLB. Antiproteolysis peptides (APPs), identified by artificial intelligence, stabilized MYC2 by binding and inhibiting PUB21 activity. A 14–amino acid peptide, APP3-14, molecularly controlled HLB in greenhouse and field trials. This approach represents a strategy to combat uncultivable pathogens through targeted disease resistance protein stabilization.

Quantifying spatiotemporal dynamics during embryogenesis is crucial for understanding congenital diseases. We developed Spateo (https://github.com/aristoteleo/spateo-release), a 3D spatiotemporal modeling framework, and applied it to a 3D mouse embryogenesis atlas at E9.5 and E11.5, capturing eight million cells. Spateo enables scalable, partial, non-rigid alignment, multi-slice refinement, and mesh correction to create molecular holograms of whole embryos. It introduces digitization methods to uncover multi-level biology from subcellular to whole organ, identifying expression gradients along orthogonal axes of emergent 3D structures, e.g., secondary organizers such as midbrain-hindbrain boundary (MHB). Spateo further jointly models intercellular and intracellular interaction to dissect signaling landscapes in 3D structures, including the zona limitans intrathalamica (ZLI). Lastly, Spateo introduces "morphometric vector fields" of cell migration and integrates spatial differential geometry to unveil molecular programs underlying asymmetrical murine heart organogenesis and others, bridging macroscopic changes with molecular dynamics. Thus, Spateo enables the study of organ ecology at a molecular level in 3D space over time.

. Tens of millions of missense variants exist in the current human population, and the vast majority of these have unknown functional consequences. Here we present a large-scale experimental analysis of human missense variants across many different proteins. Using DNA synthesis and cellular selection experiments we quantify the effect of more than 500,000 variants on the abundance of more than 500 human protein domains. This dataset reveals that 60% of pathogenic missense variants reduce protein stability. The contribution of stability to protein fitness varies across proteins and diseases and is particularly important in recessive disorders. We combine stability measurements with protein language models to annotate functional sites across proteins. Mutational effects on stability are largely conserved in homologous domains, enabling accurate stability prediction across entire protein families using energy models. Our data demonstrate the feasibility of assaying human protein variants at scale and provides a large consistent reference dataset for clinical variant interpretation and training and benchmarking of computational methods.

Human cancer cell lines have long served as tools for cancer research and drug discovery, but the presence and the source of intra-cell-line heterogeneity remain elusive. Here, we perform single-cell RNA-sequencing and ATAC-sequencing on 42 and 39 human cell lines, respectively, to illustrate both transcriptomic and epigenetic heterogeneity within individual cell lines. Our data reveal that transcriptomic heterogeneity is frequently observed in cancer cell lines of different tissue origins, often driven by multiple common transcriptional programs. Copy number variation, as well as epigenetic variation and extrachromosomal DNA distribution all contribute to the detected intra-cell-line heterogeneity. Using hypoxia treatment as an example, we demonstrate that transcriptomic heterogeneity could be reshaped by environmental stress. Overall, our study performs single-cell multi-omics of commonly used human cancer cell lines and offers mechanistic insights into the intra-cell-line heterogeneity and its dynamics, which would serve as an important resource for future cancer cell line-based studies.

BACKGROUND: The intestinal epithelial barrier confers protection against the intestinal invasion by pathogens and exposure to food antigens and toxins. Growing studies have linked the gut microbiota to the intestinal epithelial barrier function. The mining of the gut microbes that facilitate the function of intestinal epithelial barrier is urgently needed. RESULTS: Here, we studied a landscape of the gut microbiome of seven pig breeds using metagenomics and 16S rDNA gene amplicon sequencing. The results indicated an obvious difference in the gut microbiome between Congjiang miniature (CM) pigs (a native Chinese breed) and commercial Duroc × [Landrace × Yorkshire] (DLY) pigs. CM finishing pigs had stronger intestinal epithelial barrier function than the DLY finishing pigs. Fecal microbiota transplantation from CM and DLY finishing pigs to germ-free (GF) mice transferred the intestinal epithelial barrier characteristics. By comparing the gut microbiome of the recipient GF mice, we identified and validated Bacteroides fragilis as a microbial species that contributes to the intestinal epithelial barrier. B. fragilis-derived 3-phenylpropionic acid metabolite had an important function on the enhancement of intestinal epithelial barrier. Furthermore, 3-phenylpropionic acid facilitated the intestinal epithelial barrier by activating aryl hydrocarbon receptor (AhR) signaling. CONCLUSIONS: These findings suggest that manipulation of B. fragilis and 3-phenylpropionic acid is a promising strategy for improving intestinal epithelial barrier. Video Abstract.

Extensive adenosine-to-inosine (A-to-I) editing of nuclear-transcribed mRNAs is the hallmark of metazoan transcriptional regulation. Here, by profiling the RNA editomes of 22 species that cover major groups of Holozoa, we provide substantial evidence supporting A-to-I mRNA editing as a regulatory innovation originating in the last common ancestor of extant metazoans. This ancient biochemistry process is preserved in most extant metazoan phyla and primarily targets endogenous double-stranded RNA (dsRNA) formed by evolutionarily young repeats. We also find intermolecular pairing of sense-antisense transcripts as an important mechanism for forming dsRNA substrates for A-to-I editing in some but not all lineages. Likewise, recoding editing is rarely shared across lineages but preferentially targets genes involved in neural and cytoskeleton systems in bilaterians. We conclude that metazoan A-to-I editing might first emerge as a safeguard mechanism against repeat-derived dsRNA and was later co-opted into diverse biological processes due to its mutagenic nature.

Neoadjuvant chemotherapy (NAC) for rectal cancer (RC) shows promising clinical response. The modulation of the tumor microenvironment (TME) by NAC and its association with therapeutic response remain unclear. Here, we use single-cell RNA sequencing and spatial transcriptome sequencing to examine the cell dynamics in 29 patients with RC, who are sampled pairwise before and after treatment. We construct a high-resolution cellular dynamic landscape remodeled by NAC and their associations with therapeutic response. NAC markedly reshapes the populations of cancer-associated fibroblasts (CAFs), which is strongly associated with therapeutic response. The remodeled CAF subsets regulate the TME through spatial recruitment and crosstalk to activate immunity and suppress tumor progression through multiple cytokines, including CXCL12, SLIT2, and DCN. In contrast, the epithelial-mesenchymal transition of malignant cells is upregulated by CAF_FAP through MIR4435-2HG induction, resulting in worse outcomes. Our study demonstrates that NAC inhibits tumor progression and modulates the TME by remodeling CAFs.

BACKGROUND: Echinococcus tapeworms cause a severe helminthic zoonosis called echinococcosis. The genus comprises various species and genotypes, of which E. granulosus (sensu stricto) represents a significant global public health and socioeconomic burden. Mitochondrial (mt) genomes have provided useful genetic markers to explore the nature and extent of genetic diversity within Echinococcus and have underpinned phylogenetic and population structure analyses of this genus. Our recent work indicated a sequence gap (> 1 kb) in the mt genomes of E. granulosus genotype G1, which could not be determined by PCR-based Sanger sequencing. The aim of the present study was to define the complete mt genome, irrespective of structural complexities, using a long-read sequencing method. METHODS: We extracted high molecular weight genomic DNA from protoscoleces from a single cyst of E. granulosus genotype G1 from a sheep from Australia using a conventional method and sequenced it using PacBio Sequel (long-read) technology, complemented by BGISEQ-500 short-read sequencing. Sequence data obtained were assembled using a recently-developed workflow. RESULTS: We assembled a complete mt genome sequence of 17,675 bp, which is > 4 kb larger than the complete mt genomes known for E. granulosus genotype G1. This assembly includes a previously-elusive tandem repeat region, which is 4417 bp long and consists of ten near-identical 441-445 bp repeat units, each harbouring a 184 bp non-coding region and adjacent regions. We also identified a short non-coding region of 183 bp, which includes an inverted repeat. CONCLUSIONS: We report what we consider to be the first complete mt genome of E. granulosus genotype G1 and characterise all repeat regions in this genome. The numbers, sizes, sequences and functions of tandem repeat regions remain to be studied in different isolates of genotype G1 and in other genotypes and species. The discovery of such 'new' repeat elements in the mt genome of genotype G1 by PacBio sequencing raises a question about the completeness of some published genomes of taeniid cestodes assembled from conventional or short-read sequence datasets. This study shows that long-read sequencing readily overcomes the challenges of assembling repeat elements to achieve improved genomes.

SUMMARY The plant community lags far behind the animal and human fields concerning the application of single‐cell methodologies. This is primarily due to the challenges associated with plant tissue dissection and the limitations of the available technologies. However, recent advances in spatial transcriptomics enable the study of single‐cells derived from plant tissues from a spatial perspective. This technology is already successfully used to identify cell types, reconstruct cell‐fate lineages, and reveal cell‐to‐cell interactions. Future technological advancements will overcome the challenges in sample processing, data analysis, and the integration of multiple‐omics technologies. Thanks to spatial transcriptomics, we anticipate several plant research projects to significantly advance our understanding of critical aspects of plant biology.

Abstract Spatial transcriptomics measures in situ gene expression at millions of locations within a tissue 1 , hitherto with some trade-off between transcriptome depth, spatial resolution and sample size 2 . Although integration of image-based segmentation has enabled impactful work in this context, it is limited by imaging quality and tissue heterogeneity. By contrast, recent array-based technologies offer the ability to measure the entire transcriptome at subcellular resolution across large samples 3–6 . Presently, there exist no approaches for cell type identification that directly leverage this information to annotate individual cells. Here we propose a multiscale approach to automatically classify cell types at this subcellular level, using both transcriptomic information and spatial context. We showcase this on both targeted and whole-transcriptome spatial platforms, improving cell classification and morphology for human kidney tissue and pinpointing individual sparsely distributed renal mouse immune cells without reliance on image data. By integrating these predictions into a topological pipeline based on multiparameter persistent homology 7–9 , we identify cell spatial relationships characteristic of a mouse model of lupus nephritis, which we validate experimentally by immunofluorescence. The proposed framework readily generalizes to new platforms, providing a comprehensive pipeline bridging different levels of biological organization from genes through to tissues.

Colorectal cancer (CRC) patients with mismatch repair (MMR)-deficient (dMMR) but not MMR-proficient (pMMR) tend to benefit from immune checkpoint blockade (ICB) therapy. To profile the tumor microenvironments (TME) underlying these varied therapeutic responses, we integrate spatial enhanced resolution omics-sequencing (Stereo-seq), single-cell RNA sequencing, and multiplexed imaging analysis to create high-definition spatial maps of tumors from treatment-naïve and ICB-treated CRC patients. Our results identify the spatial organization and immune status of the tumor-stroma boundary as a distinctive feature of dMMR and pMMR CRCs, which associates with ICB response. The physical interactions and abundance of LAMP3+DCs and CXCL13+T cells may shape the ICB-responsive tumor-stroma boundary, whereas CXCL14+cancer-associated fibroblasts tend to remodel extracellular matrix to form a structural barrier in non-responders. Our work therefore points out the importance of the molecular and cellular spatial structures of tumors in ICB response, raising the possibility of reprogramming tumor-stroma boundary for sensitizing immunotherapies in the majority of CRCs. Colorectal cancer patients with mismatch repair-deficient, but not MMR-proficient tend to benefit from immune checkpoint blockade therapy. Here, the authors identify interactions among tumor and immune cells that are associated with the responsiveness to immunotherapy in colorectal cancer.