Cross-species cell-type assignment from single-cell RNA-seq data by a heterogeneous graph neural network

scSwinFormer: A Transformer-Based Cell-Type Annotation Method for scRNA-Seq Data Using Smooth Gene Embedding and Global Features

Single-cell omics techniques have made it possible to analyze individual cells in biological samples, providing us with a more detailed understanding of cellular heterogeneity and biological systems. Accurate identification of cell types is critical for single-cell RNA sequencing (scRNA-seq) analysis. However, scRNA-seq data are usually high dimensional and sparse, posing a great challenge to analyze scRNA-seq data. Existing cell-type annotation methods are either constrained in modeling scRNA-seq data or lack consideration of long-term dependencies of characterized genes. In this work, we developed a Transformer-based deep learning method, scSwinFormer, for the cell-type annotation of large-scale scRNA-seq data. Sequence modeling of scRNA-seq data is performed using the smooth gene embedding module, and then, the potential dependencies of genes are captured by the self-attention module. Subsequently, the global information inherent in scRNA-seq data is synthesized using the Cell Token, thereby facilitating accurate cell-type annotation. We evaluated the performance of our model against current state-of-the-art scRNA-seq cell-type annotation methods on multiple real data sets. ScSwinFormer outperforms the current state-of-the-art scRNA-seq cell-type annotation methods in both external and benchmark data set experiments.

A gene regulatory network-aware graph learning method for cell identity annotation in single-cell RNA-seq data

Cell identity annotation for single-cell transcriptome data is a crucial process for constructing cell atlases, unraveling pathogenesis, and inspiring therapeutic approaches. Currently, the efficacy of existing methodologies is contingent upon specific data sets. Nevertheless, such data are often sourced from various batches, sequencing technologies, tissues, and even species. Notably, the gene regulatory relationship remains unaffected by the aforementioned factors, highlighting the extensive gene interactions within organisms. Therefore, we propose scHGR, an automated annotation tool designed to leverage gene regulatory relationships in constructing gene-mediated cell communication graphs for single-cell transcriptome data. This strategy helps reduce noise from diverse data sources while establishing distant cellular connections, yielding valuable biological insights. Experiments involving 22 scenarios demonstrate that scHGR precisely and consistently annotates cell identities, benchmarked against state-of-the-art methods. Crucially, scHGR uncovers novel subtypes within peripheral blood mononuclear cells, specifically from CD4+ T cells and cytotoxic T cells. Furthermore, by characterizing a cell atlas comprising 56 cell types for COVID-19 patients, scHGR identifies vital factors like IL1 and calcium ions, offering insights for targeted therapeutic interventions.

Whole brain alignment of spatial transcriptomics between humans and mice with BrainAlign

The increasing utilization of mouse models in human neuroscience research places higher demands on computational methods to translate findings from the mouse brain to the human one. In this study, we develop BrainAlign, a self-supervised learning approach, for the whole brain alignment of spatial transcriptomics (ST) between humans and mice. BrainAlign encodes spots and genes simultaneously in two separated shared embedding spaces by a heterogeneous graph neural network. We demonstrate that BrainAlign could integrate cross-species spots into the embedding space and reveal the conserved brain regions supported by ST information, which facilitates the detection of homologous regions between humans and mice. Genomic analysis further presents gene expression connections between humans and mice and reveals similar expression patterns for marker genes. Moreover, BrainAlign can accurately map spatially similar homologous regions or clusters onto a unified spatial structural domain while preserving their relative positions.

Identify Regulatory eQTLs by Multiome Sequencing in Prostate Single Cells

While genome-wide association studies and expression quantitative trait loci (eQTL) analysis have made significant progress in identifying noncoding variants associated with prostate cancer risk and bulk tissue transcriptome changes, the regulatory effect of these genetic elements on gene expression remains largely unknown. Recent developments in single-cell sequencing have made it possible to perform ATAC-seq and RNA-seq profiling simultaneously to capture functional associations between chromatin accessibility and gene expression. In this study, we tested our hypothesis that this multiome single-cell approach allows for mapping regulatory elements and their target genes at prostate cancer risk loci. We applied a 10X Multiome ATAC + Gene Expression platform to encapsulate Tn5 transposase-tagged nuclei from multiple prostate cell lines for a total of 65,501 high quality single cells from RWPE1, RWPE2, PrEC, BPH1, DU145, PC3, 22Rv1 and LNCaP cell lines. To address data sparsity commonly seen in the single-cell sequencing, we performed targeted sequencing to enrich sequencing data at prostate cancer risk loci involving 2,730 candidate germline variants and 273 associated genes. Although not increasing the number of captured cells, the targeted multiome data did improve eQTL gene expression abundance by about 20% and chromatin accessibility abundance by about 5%. Based on this multiomic profiling, we further associated RNA expression alterations with chromatin accessibility of germline variants at single cell levels. Cross validation analysis showed high overlaps between the multiome associations and the bulk eQTL findings from GTEx prostate cohort. We found that about 20% of GTEx eQTLs were covered within the significant multiome associations (p-value ≤ 0.05, gene abundance percentage ≥ 5%), and roughly 10% of the multiome associations could be identified by significant GTEx eQTLs. We also analyzed accessible regions with available heterozygous SNP reads and observed more frequent association in genomic regions with allelically accessible variants (p = 0.0055). Among these findings were previously reported regulatory variants including rs60464856-RUVBL1 (multiome p-value = 0.0099 in BPH1) and rs7247241-SPINT2 (multiome p-value = 0.0002- 0.0004 in 22Rv1). We also functionally validated a new regulatory SNP and its target gene rs2474694-VPS53 (multiome p-value = 0.00956 in BPH1 and 0.00625 in DU145) by reporter assay and SILAC proteomics sequencing. Taken together, our data demonstrated the feasibility of the multiome single-cell approach for identifying regulatory SNPs and their regulated genes.

SIMS: A deep-learning label transfer tool for single-cell RNA sequencing analysis

Cell atlases serve as vital references for automating cell labeling in new samples, yet existing classification algorithms struggle with accuracy. Here we introduce SIMS (scalable, interpretable machine learning for single cell), a low-code data-efficient pipeline for single-cell RNA classification. We benchmark SIMS against datasets from different tissues and species. We demonstrate SIMS's efficacy in classifying cells in the brain, achieving high accuracy even with small training sets (<3,500 cells) and across different samples. SIMS accurately predicts neuronal subtypes in the developing brain, shedding light on genetic changes during neuronal differentiation and postmitotic fate refinement. Finally, we apply SIMS to single-cell RNA datasets of cortical organoids to predict cell identities and uncover genetic variations between cell lines. SIMS identifies cell-line differences and misannotated cell lineages in human cortical organoids derived from different pluripotent stem cell lines. Altogether, we show that SIMS is a versatile and robust tool for cell-type classification from single-cell datasets.

Single-cell phylotranscriptomics of developmental and cell type evolution

Single-cell phylotranscriptomics is an emerging tool to reveal the molecular and cellular mechanisms of evolution. We summarize its utility in studying the hourglass pattern of ontogenetic evolution and for understanding the evolutionary history of cell types. The developmental hourglass model suggests that the mid-embryonic stage is the most conserved period of development across species, which is supported by morphological and molecular studies. Single-cell phylotranscriptomic analysis has revealed previously underappreciated heterogeneity in transcriptome ages among lineages and cell types throughout development, and has identified the lineages and tissues that drive the whole-organism hourglass pattern. Single-cell transcriptome age analyses also provide important insights into the origin of germ layers, the different selective forces on tissues during adaptation, and the evolutionary relationships between cell types.

CACIMAR: cross-species analysis of cell identities, markers, regulations, and interactions using single-cell RNA sequencing data

Transcriptomic analysis across species is increasingly used to reveal conserved gene regulations which implicate crucial regulators. Cross-species analysis of single-cell RNA sequencing (scRNA-seq) data provides new opportunities to identify the cellular and molecular conservations, especially for cell types and cell type-specific gene regulations. However, few methods have been developed to analyze cross-species scRNA-seq data to uncover both molecular and cellular conservations. Here, we built a tool called CACIMAR, which can perform cross-species analysis of cell identities, markers, regulations, and interactions using scRNA-seq profiles. Based on the weighted sum models of the conserved features, we developed different conservation scores to measure the conservation of cell types, regulatory networks, and intercellular interactions. Using publicly available scRNA-seq data on retinal regeneration in mice, zebrafish, and chick, we demonstrated four main functions of CACIMAR. First, CACIMAR allows to identify conserved cell types even in evolutionarily distant species. Second, the tool facilitates the identification of evolutionarily conserved or species-specific marker genes. Third, CACIMAR enables the identification of conserved intracellular regulations, including cell type-specific regulatory subnetworks and regulators. Lastly, CACIMAR provides a unique feature for identifying conserved intercellular interactions. Overall, CACIMAR facilitates the identification of evolutionarily conserved cell types, marker genes, intracellular regulations, and intercellular interactions, providing insights into the cellular and molecular mechanisms of species evolution.

Representing and extracting knowledge from single-cell data

Single-cell analysis is currently one of the most high-resolution techniques to study biology. The large complex datasets that have been generated have spurred numerous developments in computational biology, in particular the use of advanced statistics and machine learning. This review attempts to explain the deeper theoretical concepts that underpin current state-of-the-art analysis methods. Single-cell analysis is covered from cell, through instruments, to current and upcoming models. The aim of this review is to spread concepts which are not yet in common use, especially from topology and generative processes, and how new statistical models can be developed to capture more of biology. This opens epistemological questions regarding our ontology and models, and some pointers will be given to how natural language processing (NLP) may help overcome our cognitive limitations for understanding single-cell data.

Joint representation of molecular networks from multiple species improves gene classification

Network-based machine learning (ML) has the potential for predicting novel genes associated with nearly any health and disease context. However, this approach often uses network information from only the single species under consideration even though networks for most species are noisy and incomplete. While some recent methods have begun addressing this shortcoming by using networks from more than one species, they lack one or more key desirable properties: handling networks from more than two species simultaneously, incorporating many-to-many orthology information, or generating a network representation that is reusable across different types of and newly-defined prediction tasks. Here, we present GenePlexusZoo, a framework that casts molecular networks from multiple species into a single reusable feature space for network-based ML. We demonstrate that this multi-species network representation improves both gene classification within a single species and knowledge-transfer across species, even in cases where the inter-species correspondence is undetectable based on shared orthologous genes. Thus, GenePlexusZoo enables effectively leveraging the high evolutionary molecular, functional, and phenotypic conservation across species to discover novel genes associated with diverse biological contexts.

Network models to enhance the translational impact of cross-species studies

Neuroscience studies are often carried out in animal models for the purpose of understanding specific aspects of the human condition. However, the translation of findings across species remains a substantial challenge. Network science approaches can enhance the translational impact of cross-species studies by providing a means of mapping small-scale cellular processes identified in animal model studies to larger-scale inter-regional circuits observed in humans. In this Review, we highlight the contributions of network science approaches to the development of cross-species translational research in neuroscience. We lay the foundation for our discussion by exploring the objectives of cross-species translational models. We then discuss how the development of new tools that enable the acquisition of whole-brain data in animal models with cellular resolution provides unprecedented opportunity for cross-species applications of network science approaches for understanding large-scale brain networks. We describe how these tools may support the translation of findings across species and imaging modalities and highlight future opportunities. Our overarching goal is to illustrate how the application of network science tools across human and animal model studies could deepen insight into the neurobiology that underlies phenomena observed with non-invasive neuroimaging methods and could simultaneously further our ability to translate findings across species.

Cross-species transcriptomics reveals bifurcation point during the arterial-to-hemogenic transition

Hemogenic endothelium (HE) with hematopoietic stem cell (HSC)-forming potential emerge from specialized arterial endothelial cells (AECs) undergoing the endothelial-to-hematopoietic transition (EHT) in the aorta-gonad-mesonephros (AGM) region. Characterization of this AECs subpopulation and whether this phenomenon is conserved across species remains unclear. Here we introduce HomologySeeker, a cross-species method that leverages refined mouse information to explore under-studied human EHT. Utilizing single-cell transcriptomic ensembles of EHT, HomologySeeker reveals a parallel developmental relationship between these two species, with minimal pre-HSC signals observed in human cells. The pre-HE stage contains a conserved bifurcation point between the two species, where cells progress towards HE or late AECs. By harnessing human spatial transcriptomics, we identify ligand modules that contribute to the bifurcation choice and validate CXCL12 in promoting hemogenic choice using a human in vitro differentiation system. Our findings advance human arterial-to-hemogenic transition understanding and offer valuable insights for manipulating HSC generation using in vitro models.

Cell type matching across species using protein embeddings and transfer learning

MOTIVATION

Knowing the relation between cell types is crucial for translating experimental results from mice to humans. Establishing cell type matches, however, is hindered by the biological differences between the species. A substantial amount of evolutionary information between genes that could be used to align the species is discarded by most of the current methods since they only use one-to-one orthologous genes. Some methods try to retain the information by explicitly including the relation between genes, however, not without caveats.

RESULTS

In this work, we present a model to transfer and align cell types in cross-species analysis (TACTiCS). First, TACTiCS uses a natural language processing model to match genes using their protein sequences. Next, TACTiCS employs a neural network to classify cell types within a species. Afterward, TACTiCS uses transfer learning to propagate cell type labels between species. We applied TACTiCS on scRNA-seq data of the primary motor cortex of human, mouse, and marmoset. Our model can accurately match and align cell types on these datasets. Moreover, our model outperforms Seurat and the state-of-the-art method SAMap. Finally, we show that our gene matching method results in better cell type matches than BLAST in our model.

AVAILABILITY AND IMPLEMENTATION

The implementation is available on GitHub (https://github.com/kbiharie/TACTiCS). The preprocessed datasets and trained models can be downloaded from Zenodo (https://doi.org/10.5281/zenodo.7582460).