Cattle Cell Atlas: a multi-tissue single cell expression repository for advanced bovine genomics and comparative biology

Cattle Cell Atlas: a multi-tissue single cell expression repository for advanced bovine genomics and comparative biology | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Cattle Cell Atlas: a multi-tissue single cell expression repository for advanced bovine genomics and comparative biology Lingzhao Fang, Bo Han, Houcheng Li, Qi Zhang, Weijie Zheng, Ao Chen, and 31 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4631710/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 05 Sep, 2025 Read the published version in Nature Genetics → Version 1 posted You are reading this latest preprint version Abstract Systematic characterization of the molecular states of cells in livestock tissues is essential for understanding cellular and genetic mechanisms underlying economically and ecologically important physiological traits. This knowledge contributes to the advancement of sustainable and precision agriculture-food systems. Here, as part of the Farm animal Genotype-Tissue Expression (FarmGTEx) project, we describe a comprehensive reference map comprising 1,793,854 cells from 59 bovine tissues, spanning both sexes and multiple developmental stages. This map, generated by single-cell/nucleus RNA sequencing, identifies 131 distinct cell types, revealing intra- and inter-tissue cellular heterogeneity in gene expression, transcription factor regulation, and intercellular communication. Integrative analysis with genetic variants that underpin bovine monogenic and complex traits uncovers cell types of relevance, such as spermatocytes responsible for sperm motilities and excitatory neurons for milk fat yield. Comparative analysis reveals similarities in gene expression between cattle and humans at single-cell resolution, allowing for detection of relevant cell types for studying human complex phenotypes. This cattle cell atlas will serve as a key resource for cattle genetics and genomics, immunology, comparative biology, and ultimately human biomedicine. Biological sciences/Genetics/Genomics/Transcriptomics Biological sciences/Computational biology and bioinformatics/Data processing Biological sciences/Molecular biology/Transcriptomics Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Introduction Cattle, domesticated over 10,000 years ago 1–3 , play a crucial role in converting indigestible fiber feed into protein-rich food to humans such as beef and milk, which are essential for humans nutrition and health. To meet increasing global demand for safe animal food products while minimizing the production-associated negative impacts on animal welfare and the environment (e.g., greenhouse gas emissions and zoonotic diseases) 4–7 , it is essential to understand the genetic and molecular mechanisms underlying various phenotypes of economic and ecological importance in cattle. As of 28 April, 2024, 191,181 genomic loci have been reported to be associated with over 600 different complex traits in cattle 8 . A large proportion of these variants reside in non-coding genomic regions, have small individual effects on phenotypic variation, and influence complex traits via modulation of gene regulation. In this regard, numerous studies have explored the molecular mechanisms underlying complex traits at the tissue expression levelthrough integrating genome-wide association study (GWAS) data sets and molecular quantitative trait loci (molQTL) 9–12 . For instance, the Cattle Genotype-Tissue Expression (CattleGTEx) project 13 linked gene expression in over 20 tissues with 43 significant economically important traits, provides valuable insights into their gene regulatory mechanisms. However, tissues are generally heterogenous mixtures of distinct cell types and states. As a consequence of the rapid development of single-cell sequencing technology, single-cell transcriptome atlases have been constructed for many organisms, including humans 14,15 , mice 16 , fruit flies 17 , zebrafish 18 , axolotl 19 , and pigs 20 . However, in cattle, previous studies on single-cell transcriptomic studies were limited to specific tissue types including the rumen 21–23 , peripheral blood 24,25 , skeletal muscle 26 , and the digestive system 27,28 . Therefore, there is an urgent need to comprehensively catalogue different bovine cell types and states across various tissue types and biological contexts, which will substantially contribute to our understanding of the genetic and molecular architecture underlying numerous phenotypes in cattle. In this study, we build a comprehensive Cattle Cell Atlas (CattleCA) for the livestock research community by generating and analyzing single-cell/nucleus RNA sequencing (sc/snRNA-seq) data from 1,793,854 cells across 59 tissue types in 15 animals, spanning both sexes and three developmental stages (fetus, calves, and adults; Fig. 1) . We characterize 131 distinct cell types and assess the cellular heterogeneity in terms of gene expression, transcription factor regulation, and intra- and inter-tissue cellular communications. Leveraging the CattleCA, we highlight specific cell types/states associated with monogenic conditions and complex traits. Furthermore, we explored the evolutionary conservation of the transcriptome between cattle and humans at single-cell level, revealing shared cellular mechanisms underlying complex traits and diseases in humans. The CattleCA ( https://ngdc.cncb.ac.cn/cattleca/ ) is thus of fundamental significance to serve as an invaluable resource for cattle genetics/genomics, immunology, precision breeding, and comparative biology. Results The scope of the Cattle Cell Atlas We analyzed the transcriptome data from a total of 1,506,438 single cells and 287,416 single nuclei (hereinafter referred to as cells) from one fetus, four calves, and ten adults ( Supplementary Fig. 1–1 and Supplementary Table 1–1 ). These cells included 234,802 from males and 1,559,052 from females. By integrating all these cells using Harmony 29 , we annotated 131 major cell types based on canonical marker genes, representing seven distinct cell lineages: immune (n = 679,021), endothelial (n = 308,878), epithelial (n = 268,626), stromal (n = 240,771), nerve (n = 182,450), muscle (n = 81,438), and germline (n = 4,404) cells ( Fig. 1c; Supplementary Fig. 1-2a and Supplementary Table 1–2, 3 ). These cells clustered well based on cell lineage types rather than processing methods, sequencing platforms and tissue types ( Fig. 1c and Supplementary Fig. 1-2b-d ), with cell type abundance ranging from 33 for circulating epithelial cells to 233,497 for blood vascular endothelial cells (BVECs) ( Fig. 1c ). On average, 12 distinct cell types were identified per tissue, ranging from 7 in the oviduct to 21 in the ileum ( Fig. 2a; Supplementary Table 2 − 1; and Supplementary Fig. 2 − 1 and 2–2) . Among the 131 annotated cell types, 67 were observed only in one tissue type, whereas immune, endothelial and epithelial cells were found across 58, 49, and 41 tissues, respectively ( Fig. 2a; Supplementary Fig. 1-2e and Supplementary Table 2 − 1 ). For instance, BVECs were ubiquitously present in 45 tissues, whereas α and β cells were exclusively detected in the pancreas. Furthermore, cell cycle analysis indicated that cells from germline lineage were predominantly enriched in the G2/M phase, reflecting their active growth and preparation for DNA replication. Conversely, epithelial, muscle, stromal, and endothelial cells were primarily enriched in the G0/G1 phase, indicating a quiescent state with no active division ( Supplementary Fig. 1-2f and Supplementary Table 1–4 ). Cellular heterogeneity in mammary gland and testis To study the cellular heterogeneity within tissues, we took mammary gland and testis as examples due to their potential importance in milk production and male fertility traits of economic value in dairy cattle. We identified 10 epithelial cell subtypes (ME0-ME9) from 4,819 cells isolated from mammary parenchyma, duct, and cistern ( Fig. 2b ). These cell subtypes had diverse distribution patterns and functional characteristics across anatomical components of the mammary gland ( Fig. 2c, d ). The luminal secretory (LumSec) cell subtypes (ME0, ME1, ME2, ME3, ME4, and ME5) were associated with milk biosynthesis, expressing genes like CSN2 and LALBA , and were enriched in the mammary cistern (M1, ME2, ME3, and ME4) and mammary parenchyma (ME0 and ME5; Fig. 2c, d and Supplementary Fig. 2-3a ). Genes with high expression in M1, ME2, ME3, and ME4 were significantly enriched in lipid metabolism, epithelial cell differentiation, cell morphogenesis, and immune defense, respectively ( Supplementary Fig. 2-3b ). The ME0 and ME5 subtypes, displaying upregulated genes such as CXCL8 and CCL3 ( Supplementary Fig. 2-3a ), were associated with host defense by the epithelial cells in chemotaxis and immune cell recruitment 30 . Conversely, luminal hormone-responsive (LumHR) cell subtypes, ME8 and ME9, were primarily derived from mammary ducts ( Fig. 2c ) and involved in ductal development 31 through paracrine EGFR-ligands, AREG , TGFA , BTC , and HBEGF ( Supplementary Fig. 2-3c ), cross-acting on myoepithelial cells (ME7). In testis, we categorized 3,958 germline cells into 12 cell subtypes, representing the progression from spermatogonia (S5 and S7), through spermatocytes (S0, S3, S6, S9, and S10), to spermatids (S1, S2, S4, S8, and S11; Fig. 2e and Supplementary Fig. 2–4 ). Pseudotime analysis validated the developmental trajectory of gametogenesis ( Fig. 2f ). Interestingly, spermatocytes displayed a bifurcated differentiation pattern from the early stage (S3) to the late stage (S0) ( Fig. 2f ), consistent with previous findings in yaks or cattle-yak hybrids 32 , but not in humans 33 . Furthermore, we identified nine pivotal transcription factors (TFs) orchestrating the gametogenesis process ( Fig. 2g). For instance, BCLAF1 and HLTF exhibited reduced expression levels during the spermatogonial transition (S5 to S7), suggesting putative roles for these TFs in sustaining spermatogonial stemness. YY1, known to influence chromosome double-strand breaks 34 , exhibited a peak expression in S3. CREM and MBD1 demonstrated increased expression from late spermatocytes onwards, which might influence sperm development 35,36 . As crucial TFs for spermiogenesis, the expression patterns of ELF2, RFX2, RFX4, and CUX1 remained constant during the process from late spermatogonia to the maturation of spermatids ( Supplementary Fig. 2–5 ). Intercellular communication and transcriptional regulation within tissues Cellular interactions play essential roles in shaping the functions of multicellular organisms 37 . Using CellChat 38 , we found that most cellular communications occurred within cell lineages, with immune and epithelial cell lineages showing higher communication ( Supplementary Fig. 3 − 1 ). Stromal cells emitted the most signals, essential for maintaining tissue homeostasis ( Supplementary Fig. 3 − 1 ). The intensity and pattern of intercellular interactions were tissue-specific, with the uterine horn showing the strongest and the retina the weakest interactions ( Fig. 3a ). To dissect gene regulatory networks at the cell type level, we conducted a systematic analysis using pySCENIC 39 , identifying 14,094 regulons for 980 TFs across 59 tissues ( Supplementary Table 3 − 1 ). Among these TFs, 710 showed cell-type-specific activity across 130 cell types (excluding proliferative cells), potentially playing pivotal roles in determining cellular identity and fate. For example, GATA1, SPI, and HNF4A are crucial for mast cell, macrophage, and hepatocyte differentiation, respectively ( Fig. 3b ). TFs regulating a broader spectrum of cell types exhibited stronger DNA sequence constraints ( Supplementary Fig. 3-2a, b ). For instance, HOXA2, which influences regulatory processes across 35 distinct cell types, had the highest PhastCons and phyloP scores, within its target genes which were significantly enriched in the cytoplasm, negative regulation of expression, and identical protein binding ( Supplementary Fig. 3-2a-c ). To investigate the impact of sex on transcriptional regulation, we compared male and female regulon activity scores (RAS) across 19 tissues with data for both sexes ( Fig. 3c; Supplementary Fig. 1–1 and Supplementary Fig. 3–3 ). We identified one tissue exhibiting no sex-biased regulation (medulla oblongata), four tissues displaying sex-biased regulation (i.e., heart, kidney, esophagus, and skin), and the remaining 14 tissues manifesting partial sex-biased regulation limited to specific cell types ( Fig. 3c and Supplementary Fig. 3–3 ). Taking the liver as an example, three TFs, HNF4A, NR1I3, and NFIA, exhibited high regulatory activity specifically in female hepatocytes, with their target genes (e.g., AGMO, KHK , and FTCD ; Fig. 3d ) participating in fatty acid metabolism 40–42 . Similarly, SOX18 specifically regulates female liver sinusoidal endothelial cells (LSECs) and BVECs, targeting signaling pathway-related genes 43,44 , such as HPCAL1 and RAMP3 ( Fig. 3d ). The four TFs and their five target genes detailed above also have higher expression in the female human liver compared to the male liver ( https://gtexportal.org/home/ ). We further analyzed the cellular communication between hepatocytes and the two endothelial cells in females and found that their ligand-receptor pairs primarily participate in the VEGF signaling pathway, including VEGFA-VEGFR1, VEGFA-VEGFR1R2, and VEGFA-VEGFR2 ( Fig. 3e ). The VEGF pathway plays essential roles in influx and efflux of substances such as lipoproteins and chylomicron remnants 45,46 . Taken together, our findings indicate that these sex-biased regulatory factors in the liver may play key roles in lipid metabolism and milk production in lactating dairy cows. Antigen-presenting immune cell heterogeneity We identified 777,873 immune cells from 58 tissues (excluding the retina), categorizing them into 29 cell types ( Fig. 4a and Supplementary Table 4 − 1 ). The mammary gland (96,865), liver (55,369), and PBMC (46,078) exhibited the highest number of immune cells. While macrophages were found across 38 tissues, double-negative T cells, Kupffer cells, and plasmacytoid dendritic cells (pDCs) were detected only in thymus, liver, and PBMC, respectively ( Fig. 4a ). According to gene expression similarities, lymphocytes and myeloid cells (MCs) were well-clustered, indicating their similar functional and lineage-specific features, whereas microglia, neutrophils, and pDCs displayed a weak correlation with other immune cells ( Fig. 4b ). Plasma cells and erythroid cells showed pronounced heterogeneity compared to other immune cells ( Fig. 4b ). Antigen-presenting cells (APCs) are central to the induction of adaptive immunity, mediating immune responses by processing and presenting antigens to lymphocytes 47,48 . Within the myeloid cell cluster, a key component of APCs, we identified 169,610 cells from 44 tissues, which were further categorized into six macrophage subtypes (MA0-MA5), three monocyte subtypes (MO0-MO2), and four dendritic cell subtypes (DC0-DC3) ( Fig. 4c and Supplementary Fig. 4-1a ). A total of 135,186 macrophages were present in 40 tissues, making macrophages an appropriate model for exploring cellular heterogeneity across tissue microenvironments. We found the heterogeneity for the following aspects: 1) unique transcription factor regulation was observed across cell subtypes ( Fig. 4d ); elevated expression levels of IRF7 and IRF8 in MA1 potentially sustain macrophage functionality and development, while BHLHE41 , highly expressed in MA0, may inhibit the replacement of exogenous macrophages and enhance tissue-specific macrophage characteristics; 2) varied antigen-presenting ability was evident among cell subtypes, exemplified by MA1, primarily distributed in the intestine, exhibiting the strongest MHC-II antigen-presenting score (APS), contrasting with MA4 in the liver, which displayed the weakest APS ( Supplementary Fig. 4-1b, c and Supplementary Table 4 − 2 ); 3) a Meta-neighbor analysis revealed six functional modules ( Supplementary Fig. 4-2a, c-j ), where Module 6, characterized by upregulated genes associated with intracellular signal transduction and primarily composed of MA0 from the brain, accounted for 92.4% of the cells in that tissue ( Supplementary Fig. 4-2a, b ); 4) across 20 tissues, macrophages exhibited a propensity towards the classically activated M1 phenotype, predominantly in the reproductive and nervous systems; conversely, 19 tissues exhibited a tendency towards alternatively activated M2 macrophages which was observed in tissues such as the intestine ( Supplementary Fig. 4-1e ); 5) pseudotime trajectory analysis revealed a bifurcated differentiation pattern of macrophages derived from monocytes ( Fig. 4e ); on the right branch, MA1 and MA2, derived from more developed monocytes (MO2), exhibited activated genes related to transmembrane transport. Conversely, the left branch depicted gradual differentiation of MA4, MA3, and MA0, with activated genes enriched in neural development, including microglia markers such as P2RY12 , GPC5 , CALCR , and CX3CR1 ( Fig. 4e-h and Supplementary Fig. 4-1d ). Since macrophages can be derived from either circulating monocytes or embryonic progenitor cells 49–51 , it is suggested that MA1 and MA2 might be derived from monocytes, while MA0 (brain accounting for 96.24%), MA3 (64.46% liver, 30.58% spleen) and MA4 (96.24% liver) with tissue specificity might be derived from embryonic progenitor cells. Similarly, we investigated the heterogeneity of B cells across different tissues, as they can also act as APCs 52,53 . We categorized 76,410 B cells from 29 tissues into 8 subtypes ( Supplementary Fig. 4-3a-c) . Transcripts for key TFs associated with diverse B cell differentiation fates, such as YY1, BCL6, and BACH2, were detected ( Supplementary Fig. 4-3d ), which play crucial roles in developing germinal center B cells 54,55 . ATF4 and XBP1 , highly expressed in both P0 and P2, were found to regulate B cell differentiation into plasma cells 56,57 . Plasma cells exhibited a significantly lower APS for MHC-II compared to B cells, which may be attributed to their origin from antigen-stimulated B cells and their primary antibody production function and secretion rather than antigen presentation 58 ( Supplementary Fig. 4-3e ). Pseudotime trajectory analysis demonstrated the differentiation process from B cells to plasma cells ( Supplementary Fig. 4-3f, g ). During this process, markers specific to B0/B1 subtypes, such as EBF1, BACH2 , and CCR7 , were upregulated in the early stages of differentiation, followed by the activation of intermediate B cell subtype markers, including CD79A , CD79B , BCL6 , and NEIL1 . Finally, the expression of plasma cell markers like JCHAIN and MZB1 rapidly increased during differentiation into plasma cells ( Supplementary Fig. 4-3h, i ). Epithelial cell heterogeneity and their interactions with immune cells in the intestine We analyzed 278,584 epithelial cells across41 tissues, identifying 50 subtypes ( Supplementary Fig. 5-1a and Supplementary Table 5 − 1 ). Although most subtypes were tissue-specific, keratinocytes ( 36 , 114 ), spinous cells (22,672), basal cells (19,283), and goblet cells (13,055) exhibited the highest cell counts and were identified in more than six tissues ( Supplementary Fig. 5-1a ). Cell types with similar biological functions were clustered together, such as chief cells, parietal cells, isthmus cells, mucous neck cells, and pit cells, all of which have the function of secreting gastric protease and promoting digestion ( Supplementary Fig. 5-1b ). Additionally, three regions of the forestomach (rumen, reticulum, and omasum) and six regions of the intestine (duodenum, ileum, jejunum, colon, cecum, and rectum) were also clustered together ( Supplementary Fig. 5-1c ). Epithelial cells play pivotal roles in absorption, metabolism, and immune defense 59,60 , making a thorough investigation of their functional and regulatory diversity essential, particularly within the digestive system. We scrutinized 156,832 epithelial cells across 15 digestive tissues, uncovering eight cell-to-tissue modules with distinct biological functions and TF regulation ( Fig. 5a ). For instance, epithelial cells in the rumen, reticulum, and omasum exhibited upregulated genes enriched in amino acid and fatty acid degradation, regulated by TFs such as OVOL2 and PITX1 ( Fig. 5a ). In comparison with the three ruminant forestomach compartments, the cell types of the human stomach and abomasum are nearly identical, and the TF regulation pattern of their epithelial cells is analogous. The highly expressed genes in the stomach and the abomasum are predominantly enriched in protein processing and amino acid processing, which is important for secreting gastric protease to facilitate digestion ( Supplementary Fig. 5-2a-d ). In addition, the high-expression genes of the stomach were enriched in pyruvate and sugar metabolism, suggesting that the human stomach also has a similar function to the cattle forestomach ( Supplementary Fig. 5-2c ). The developmental lineage of the forestomach and abomasum were clearly separated ( Supplementary Fig. 5-2e ), supporting the multi-origin hypothesis that the forestomach originates from the esophagus while the abomasum originates from duodenum 61 . Genes involved in gastric acid secretion and protein processing were significantly activated during the development of the abomasum ( Supplementary Fig. 5-2f, g ). To explore the heterogeneity of spinous cells in the forestomach, we further divided them into 13 modules based on their gene co-expression patterns ( Supplementary Fig. 5-3a, b ). Modules predominated in the rumen showing upregulated genes associated with fatty acid oxidation and proton transport ( Supplementary Fig. 5-3c, d ), regulated by TFs including PITX1, HMGA1, HOXC6, and YBX1 ( Supplementary Fig. 5-3e ). These findings underscore the crucial role of rumen spinous cells in fatty acid absorption. In addition, hepatocytes (HEPs) in the liver and bile duct exhibited upregulation of genes enriched in bile secretion and energy metabolism pathways, with specific TFs such as NR1H4 for bile salt synthesis, ONECUT2 for hepatocyte differentiation ( Fig. 5a ), and HNF4A and NFIA for fat, protein, and sugar metabolism ( Supplementary Fig. 5-4a-b ). HEPs were further subdivided into three distinct subtypes (HEP0-HEP2; Supplementary Fig. 5-4c ), with HEP0 associated with bile secretion and VFA metabolism, HEP1 with glycolysis, and HEP2 with the calcium signaling pathway ( Supplementary Fig. 5-4d, e) . Epithelial cells in the intestine were characterized by upregulated genes involved in fat and mineral digestion and absorption under the regulation of CDX1 and CDX2, two TFs implicated in intestinal cell differentiation and inflammation 62,63 ( Fig. 5a ). Given the pivotal role of goblet cells (GCs) in maintaining the intestinal immune barrier and inflammatory response 64 , we investigated cellular communication between GCs and seven types of immune cells. Interestingly, GCs had significantly stronger communication with conventional dendritic cells type 1 (cDC1) and macrophages in the ileum and cecum than in other intestinal tissues ( Supplementary Fig. 5-5a, b ). Most of the ligand-receptor pairs were shared between GCs and immune cells, with the APP-CD74 pair showing the highest communication probability, in which CD74 signaling transduction is strongly activated during intestinal inflammation and protects the host by promoting epithelial cell regeneration, healing, and maintaining mucosal barrier integrity 65 ( Fig. 5b-d ). Significant cellular interactions among GCs were identified through CDH1-CDH1 pairs, which potentially maintain the intestinal barrier and have been shown to be important in the pathogenesis of ulcerative colitis and Crohn’s disease (CD) 66,67 . Moreover, GCs were further categorized into seven subtypes with distinct biological functions (GC0-GC6; Fig. 5e, f ). For example, genes upregulated in GC1 were enriched in intracellular signal transduction, whereas those in GC4 were enriched in regulation of immune system processes, and exhibit the strongest cellular communication strength between GC1 and GC4 ( Fig. 5f, g ). Among the 128 identified significant cellular communications among GC subtypes and immune cells ( Supplementary Table 5 − 2 ), GC1 exhibited the strongest interactions with GC4 through the CDH signaling pathway, while GC5 and GC6 displayed no interplay with other GCs ( Fig. 5h ). GCs had stronger communication with macrophages (MA1) and cDC1 compared to other immune cells, predominantly via the APP signaling pathway, with GC1 and GC4 showing the strongest interactions ( Fig. 5i ). Cellular basis and mechanisms underlying monogenic disorders in cattle To explore whether the CattleCA could serve as a powerful resource for the dissection of the cellular basis and mechanisms underlying monogenic conditions in cattle, we compiled 183 causal genes associated with 145 bovine disorders from the Online Mendelian Inheritance in Animals (OMIA) database 68 and divided them into 10 trait domains based on their phenotypic manifestations ( Supplementary Table 6 − 1) . We detected 2,677 cell type-specific genes (z-score > 0.75) across 129 distinct cell types (excluding PVALB + GABAergic neurons and proliferative cells) ( Fig. 6a and Supplementary Table 6 − 2) . Our enrichment analysis revealed significant overlaps between cell lineage-specific genes and causal genes of disorders. For instance, muscle cell-specific genes were significantly enriched with muscle disorder genes (FDR < 0.05), germline cells with reproduction disorders, and epithelial cells with skin disorders ( Fig. 6b ). Nine out of 27 genes related to skin disorders exhibited significantly higher expression in epithelial cells compared to other cell lineages ( Fig. 6c and Supplementary Fig. 6-1a ). Disorder-related genes, such as DSP for ichthyosis and TSR2 for hypotrichosis showed ubiquitous expression across various types of epithelial cells ( Fig. 6d ). DSP , encoding the desmoplakin protein, plays a pivotal role in maintaining functional desmosomes crucial for skin cell integrity and adhesion 69 , while TSR2 is primarily involved in apoptosis 70 . Six other genes with pivotal roles in pigmentation, membrane production, and fat/enzyme transport, including KRT5 , LAMA3 , and LAMC2 for epidermolysis bullosa (EB), ABCA12 for ichthyosis, MLPH for coat color, and PRLR for slick hair 71–73 , displayed specific expression in duct and basal cells. Additionally, SLC45A2 , related to coat coloration, exhibited specific expression in melanocytes, and has previously been implicated in human melanin production 74 . To explore the cell-cell interaction of nine epithelial-specifically expressed disease genes, we performed cellular communication analysis, identifying laminin, encoded by LAMA3 and LAMC2 , as a key ligand in uninucleate trophoblast cells (UTCs) (found in the placenta) and progenitor cells (present in the jejunum and duodenum), respectively ( Fig. 6e and Supplementary Fig. 6 − 2 ). LAMA3 showed high expression in placental uninucleate trophoblast cells, while LAMC2 exhibited elevated expression in intestinal progenitor cells ( Fig. 6f and Supplementary Fig. 6-1b ). LAMA3 and LAMC2 play crucial roles in regulating skin strength and resiliency 75 , with mutations disrupting laminin assembly and leading to EB in cattle 76 . A previous study has also reported that laminin deficiency can impact trophoblast differentiation and embryonic development 77 . Trajectory analysis of UTCs revealed specific expression of LAMA3 towards the end of UTC differentiation ( Fig. 6g and Supplementary Fig. 6-1c ). Furthermore, we annotated UTCs into seven subtypes (UTC0-UTC6; Supplementary Fig. 6-1d ). Among them, LAMA3 was predominantly expressed in the UTC2 and UTC3 ( Fig. 6g and Supplementary Fig. 6-1c ). Notably, UTC2 and UTC3, in which LAMA3 plays an important role as a ligand-encoding gene, exhibited stronger cellular communication with other cell types within the placenta compared to other five subtypes, indicating a potential correlation between EB and UTC2/UTC3 ( Fig. 6h; Supplementary Fig. 6-1e; and Supplementary Fig. 6 − 3 ). Furthermore, marker genes of UTC2 and UTC3 were significantly enriched in several membrane pathways, including the basement membrane, the sarcolemma, and the apical plasma membrane ( Supplementary Fig. 6-1f ). These pathways regulate cell-cell adhesion and affect the structure and stability of the basement membrane zone, which is the key pathogenic mechanism underlying EB 78 . In addition, LAMA3 , positioned within these pathways, provides a compelling hypothesis that LAMA3 , by participating in membrane function as a ligand-encoding gene within UTC2 and UTC3, may potentially contribute to EB in cattle ( Supplementary Table 6 − 3) . For muscle disorders, all 11 identified causal genes exhibited specific expression patterns (z score > 0.75) within distinct muscle cells ( Fig. 6i ). Among them, MYBPC1 and PPP1R13L displayed upregulated expression in skeletal muscle cells within the esophagus and cardiomyocytes in the heart, respectively ( Fig. 6j and Supplementary Fig. 6-4a ). MYBPC1 , identified as a candidate gene associated with increased muscular tonus (IMT) in both humans and cattle 79,80 , encodes the type I myonuclear isoform of myosin-binding protein C, an essential regulator of muscle contraction and tonus in humans 81 , therby underscoring a significant relationship between skeletal muscle cells and IMT. On the other hand, PPP1R13L , encoding the inhibitor of apoptosis-stimulating p53 protein (iASPP), plays regulatory and inflammatory roles in cardiomyocytes, potentially inducing cardiomyopathy and woolly haircoat syndrome (CMWH) in dairy cattle 82 . We further classified skeletal muscle cells (SMCs) into five subtypes (SMC0-SMC4), and trajectory analysis revealed specific expression of MYBPC1 in SMC0 and at the late stage of skeletal muscle cell differentiation, corresponding to type I myonuclear cells ( Fig. 6k; Supplementary Fig. 6-4b, c ). Marker genes of SMC0 were predominantly enriched in muscle contraction, actin filament binding, and the Z disc (Supplementary Fig. 6-4d ), which were highly related to muscle tonus 83 . In addition, marker genes in SMC4 exhibited significant enrichment in immune response pathways, suggesting that skeletal muscle cells might have certain immune regulatory functions at the onset of differentiation ( Fig. 6k and Supplementary Fig. 6-4d ). Furthermore, we examined the expression patterns of four genes associated with reproductive disorders across four germline cell types and found that ABHD16B and TMEM95 exhibited upregulated expression in spermatids compared to other germline cells, associated with infertility and male subfertility, respectively ( Supplementary Fig. 6-4e ). Additionally, specifically high expression of three blood/immune disorder genes, ITGB2 , F13A1 , and RASGRP2 , was observed in immune cells ( Supplementary Fig. 6-4f, g ). Cellular basis and mechanisms underlying complex traits in cattle To explore whether the CattleCA could contribute to unraveling some of the cellular basis and mechanisms underlying complex traits in cattle, we collected GWAS summary statistics for 55 complex traits, representing milk production (n = 30), male fertility (n = 5), coat color (n = 6), immunoglobulin G (IgG) (n = 10), body conformation (n = 3) and health traits (n = 1). To prioritize cell types involved in these complex traits, we conducted a trait-cell type enrichment analysis using scPagwas 84 , and revealed certain associations between cell lineages and complex traits. For instance, nerve cells were associated with milk production and germline cells with sperm traits ( Fig. 7a and Supplementary Fig. 7 − 1 ). For milk production traits, the most significantly associated cell types were neurons. Excitatory neurons, specifically, exhibited significant associations with milk fat yield (FY; p -value = 8.25×10 − 19 ) and capric acid (C10:0; p -value = 1.51×10 − 20 ), alongside amacrine cells displaying associations with lauric acid (C12:0; p -value = 4.3×10 − 16 ; Fig. 7b ). This finding is consistent with our previous observation at the bulk tissue level in cattle, where a strong association between neurobiology and milk production traits was noted 85 , and aligns with previous research in humans suggesting reciprocal regulation between neuronal activity and lipid metabolism 86–88 . Additionally, we observed that excitatory neurons in the cerebral cortex and amacrine cells in the retina exhibited significant associations and high trait-relevant scores (TRSs) for FY ( p -value = 8.87×10 − 6 ), C10:0 ( p -value = 8.75×10 − 6 ), and C12:0 ( p -value = 1.59×10 − 2 ) as compared to other cell types, underscoring the important roles of these two tissues in regulating milk fat content ( Fig. 7d and Supplementary Fig. 7-2a, b) . Furthermore, skeletal muscle cells showed a significant association with C12:0 ( p -value = 9.14×10 − 20 ; Fig. 7b ), consistent with their pivotal role in fatty acid utilization and intracellular fatty acid homeostasis 89,90 . Further analysis revealed that skeletal muscle cells in the esophagus ( p -value = 6.89×10 − 13 ) and tongue ( p -value = 1.03×10 − 12 ) showed significant associations with C12:0, indicating a prominent role for these two tissues in fatty acid regulation ( Fig. 7e and Supplementary Fig. 7-2c ). According to the pathway activity analysis conducted with scPagwas, neuronal excitement and lipolysis related pathways, like thermogenesis ( p -value = 1.15×10 − 198 ), regulation of lipolysis in adipocytes ( p -value = 2.07×10 − 207 ) and glutamatergic synapse ( p -value = 2×10 − 197 ), had significant activity in excitatory neurons, amacrine cells and skeletal muscle cells respectively, and the genes in these pathways had significant enrichment with fatty acid trait-relevant genes ( Fig. 7g, Supplementary Fig. 7-2f and Supplementary Table 7 − 1, 2 ). In addition, we observed that luminal cells, closely related to lactation and milk production traits 91 , exhibited a particularly significant association with the production of pentadecanoic acid (C15:0) ( Fig. 7b ), along with active involvement in pathways regulating lipolysis in adipocytes ( p -value = 2.11×10 − 218 ), PI3K-Akt ( p -value = 1.73×10 − 292 ), and retrograde endocannabinoid signaling pathways ( p -value = 3.2×10 − 286 )( Fig. 7g ). For male fertility traits, the most significant association was observed between sperm motilities (SMOT) and spermatocytes across 129 cell types ( p -value = 1.11×10 − 7 ; Fig. 7c ). Spermatocytes undergo a complex process of differentiation following meiosis, ultimately maturing into spermatids. This is a crucial step that directly impacts the quality of sperm produced 92 . This finding was further reinforced by the significant association observed between spermatocytes in the testis and SMOT ( p -value = 2.83×10 − 10 ) within the testis tissue level, with the function of active pathways in spermatocytes primarily focused on nutrient metabolism ( Fig. 7f, g ). These pathways also had significant enrichment with SMOT-relevant genes of spermatocytes, indicating potential association of them with sperm production and maturation ( Fig. 7g and Supplementary Table 7 − 2 ). Additionally, pDCs derived from PBMC were significantly associated with semen concentration per ejaculate (SCPE) trait ( p -value = 2.06×10 − 5 in the global atlas; p -value = 7.31×10 − 4 in the PBMC; Fig. 7c and Supplementary Fig. 7-2e ), consistent with our previous findings regarding the involvement of immune cells in male fertility traits 85 . This aligns with reports in humans indicating a significant association between DC abundance and sperm quality, suggesting a potential contribution of DC-mediated immune responses suboptimal to male fertility or infertility 93 . Pathway analysis revealed significant activity in distal convoluted tubule cells predominantly involving the regulation of hypothalamic gonadotropin-releasing hormone (GnRH) signaling pathway ( p -value = 8.83×10 − 250 ; Fig. 7g ). Furthermore, signal transduction pathways, like dopaminergic synapse, had significant activity in cone photoreceptor cells ( p -value = 5.59×10 − 265 ; Fig. 7g ). Additionally, the cell types most associated with coat color were type B intercalated cells, luminal cells, and amacrine cells; for IgG levels, they were type B intercalated cells, CD8 + T cells, and skeletal muscle cells; and for body and health, they were chondrocytes, epithelial stem cells, and amacrine cells ( Supplementary Fig. 7-3a-c ). In summary, the association of specific cell types with complex traits provides a cellular perspective for pinpointing the genetic regulatory mechanisms underlying important cattle traits. Cross-species cell transcriptome similarity comparisons and disease associations To investigate the similarities in gene expression, TF regulation, and cellular communication between cattle and humans at the single-cell level, we collected and analyzed 18 publicly available single-cell transcriptome datasets from 30 human tissues ( Supplementary Table 8 − 1 ). Using manual cell type annotation approach, we identified 106 distinct cell types, with 68 shared between cattle and humans. The Meta-neighbor analysis revealed conservation in gene expression between cattle and humans, particularly in nerve and immune cells (AUROC > 0.90; Fig. 8a ). Furthermore, the highly correlated cell-type regulon specificity scores (RSSs) of orthologous TFs ( r = 0.52–0.77) also showed the evolutionary conservation of TF regulation, particularly in immune and epithelial cells ( Fig. 8b, c; Supplementary Table 8 − 2 and Supplementary Fig. 8 − 1 ). For example, TFs crucial for B lymphocyte function, such as REL, ETS1, and ARID3A, exhibited RSS exceeding 0.25 in immune cells across both species, suggesting a conserved regulatory function in immune responses. Similarly, TFs such as HNF4G, CDX2, EHF, KLF4, and PITX1, implicated in epithelial cell differentiation and the establishment of the epithelial barrier 94–98 , also had RSSs over 0.25 in both cattle and human epithelial cells, suggesting analogous regulatory mechanisms governing epithelial function. Furthermore, we observed notable similarities in cellular communication and signaling pathways, particularly in tissues like the small intestine and liver, between the two species ( Fig. 8d-g and Supplementary Fig. 8 ). For instance, in the small intestine of the two species, enterocytes, goblet cells, Schwann cells, and B cells showed similar communication strengths and shared signaling pathways. In the liver, Kuffer cells, hepatocytes, and cholangiocytes demonstrated similar communication strength between the two species. However, several tissues, such as the mammary gland ( Supplementary Fig. 8 − 2 ), cerebellum ( Supplementary Fig. 8 − 2) , esophagus ( Supplementary Fig. 8 − 3) , and retina ( Supplementary Fig. 8 − 4) , displayed distinct cellular communication patterns between species. These findings together indicate considerable conservation at the cellular expression, TF regulation, and communication between cattle and humans, providing valuable insights into cross-species similarities. To explore whether the CattleCA resource can contribute to the explanation of genetic and cellular mechanisms underlying complex human traits and diseases, we analyzed the heritability enrichment of 43 human traits and diseases by linkage disequilibrium score regression (LDSC) on the orthologous marker genes across 49 cattle tissues. Significant enrichment of heritability for human complex traits and diseases was found in corresponding tissues in cattle ( Supplementary Table 8 − 3 and Supplementary Fig. 8-6a ). For instance, the orthologous marker genes in cattle jejunum, colon, and reticulum showed significant enrichment of heritability for human inflammatory bowel disease (IBD) and CD, while markers in the blood, cecum, ileum, and pituitary gland showed enrichment for multiple sclerosis (MS) (FDR < 0.05). We analyzed four diseases - MS, IBD, RA, and CD, due to their significant heritability enrichment and the potential links in their mechanisms of onset and progression in humans and cattle 99,100 . We calculated the heritability enrichment for these conditions across seven tissues by extending the categories of orthologous marker genes to cell types. Remarkably, certain cell types showed significant heritability enrichments, enhancing our understanding of their genetic and molecular basis. For instance, CD4 + T cells exhibited substantial heritability enrichment for MS and RA, while CD8 + T cells showed enrichment for CD and IBD ( Fig. 8h-i ; Supplementary Fig. 8-6b-c and Supplementary Table 8 − 3 ). Moreover, several cell types displayed significant heritability enrichments for diseases have not been extensively studied. For example, microglia in the pituitary gland showed significant enrichment for MS ( Fig. 8h ), and the differentially expressed genes in pituitary microglia were enriched in signal transduction by p53 class mediator ( Supplementary Fig. 4-2j ), suggesting that activation of the apoptotic protein p53 be associated with MS progression. Additionally, GCs in the colon were significantly enriched for IBD ( Fig. 8i ) and had strong communication with immune cells ( Fig. 5b-d ), suggesting their potential protective roles in the intestinal epithelial barrier. Discussion In summary, through generating and analyzing gene expression of 1,793,854 cells across 59 tissue types in 15 cattle, we annotated 131 cell types to build the first version of CattleCA and provided a web portal for the community to explore and query all the results. The following comprehensive inter- and intra-tissue analyses advanced our understanding of cellular heterogeneity across the cattle body and between sexes. Integrating this resource with large scale population genetics data facilitated the detection of relevant cell types and stages for complex phenotypes. For instance, among all the annotated cell types, spermatocytes and excitatory neurons showed the strongest association with sperm motilities (SMOT) and milk fat yield, respectively, in dairy cattle. This knowledge can further sever as biological priors for prioritizing causal genes and variants as well as improving the prediction accuracy of complex phenotypes 85,101 . The cross-species comparative analysis revealed a high similarity in gene expression, regulation, and cellular communication between cattle and humans at single cell resolution. These findings will contribute to understanding the cellular and evolutionary mechanisms underlying zoonoses associated with cattle, such as tuberculosis, salmonellosis and ringworm 102 , facilitating the development of bovine models for certain human diseases 103,104 . Although it provides a valuable resource for the cattle genetics and genomics community, the current CattleCA still has certain limitations: 1) Our current samples were limited to Holstein cattle, a dairy breed of global economic value. It is thus imperative to encompass a broader range of cattle breeds, such as beef cattle. 2) More biological and environmental contexts should be considered in the future development of CattleCA, such as embryonic stages, healthy status, and diet changes, because some cell types and states may only be observed in certain contexts. 3) Additional single-cell omics data, such as epigenome, proteome, and spatial transcriptome, are required for accurately and precisely annotating cell types and stages 105 . A more comprehensive CattleCA will enable a deeper understanding of the dynamic landscape of cellular function across diverse biological and environmental contexts, advancing our understanding of molecular mechanisms underlying complex phenotypes and environmental adaptations in cattle and even in humans. Methods Ethics statement The experimental procedures were approved by the Institutional Animal Care and Use Committee (IACUC) at China Agricultural University (Beijing, China; approval number: DK996) and Northwest A&F University (Yangling, China; approval number: DK20230113), and the Animal Care Committee of Zhejiang University (Hangzhou, China; approval number ZJU202017326). Sample and data collection Two healthy Chinese Holstein cows, sharing the same sire, similar birth and calving dates, and lactation performance, were selected from Beijing Sunlon Livestock Development Co., Ltd. (Beijing, China) were used. Both cows were in the mid to late stages of their first lactation. Following a 12-hr fasting period, animals were euthanized and 41 tissue samples were collected from each cow (see individuals 1 and 2 in Supplementary Fig. 1–1 ). The tissue samples were immediately placed into sterile serum-free Minimum Essential Medium (MEM) at 4°C and subsequently transported to the tissue culture laboratory for single-cell/nucleus suspension preparation within 4 hr. In addition, we also obtained single-cell/nucleus transcriptome data of Holstein cattle tissues from our collaborative teams 22,27 and public databases (rumen 21 , ovary 106 , placenta 107 , and intervertebral discs 108 ). Overall, this study analyzed a total of 152 samples representing 59 tissues with one to nine replicates per tissue (including 10 tissues from different anatomical structures of tissues/organs, namely mammary parenchyma, mammary duct, mammary cistern, rumen papilla, rumen muscle, frontal lobe, occipital lobe, parietal lobe, circumvallate papilla, and fungal papilla) from 11 systems (comprising the digestive system, endocrine system, nervous system, cardiovascular system, respiratory system, urinary system, lymphatic system, female reproductive system, male reproductive system, skeletal system, and integumentary system) collected from 15 Holstein cattle (encompassing nine adult cows, one adult bull, one weaned female calf, three one-day-old female calf, and one female fetus) ( Fig. 1b; Supplementary Fig. 1–1 and Supplementary Table 1–1 ). We also used single-cell transcriptome data from 30 human tissues for interspecies analysis ( Supplementary Table 8 − 1 ). Single-cell suspension preparation Bone marrow : The fresh marrow form leg bone was minced and suspended i n 5 mL of RPMI 1640 cell culture medium (Gibco). The cells were filtered through a 40 µm SmartStrainer (Miltenyi Biotec), centrifuged at 300 ×g for 10 min at 4℃, and resuspended in 1 mL of 1× phosphate buffered saline (PBS). The cell suspension was mixed with Red Cell Lysis Solution (Solarbio) at a cell to lysis buffer ratio of 1:3, allowed to stand for 4–8 min, centrifuged at 300 ×g for 5 min at 4℃, and resuspended in 1×PBS with 0.04% bovine serum albumin (BSA). Cervical lymph node and adrenal gland : The fresh tissues were washed with 1×PBS, cut into 2 mm pieces, and incubated with dissociation enzymes according to manufacturer’s instructions (Tumor Dissociation Kit, Miltenyi Biotec) at 37℃ for 30–60 min with gentle rotation. Then, the cell suspension was filtered through a 40 µm SmartStrainer, centrifuged at 300 ×g for 10 min at 4℃, and resuspended in 1 mL of 1×PBS. The cell suspension was mixed with Red Cell Lysis Solution at a ratio of 1:3, allowed to stand for 3–5 min, centrifuged at 300 ×g for 5 min at 4℃, and resuspended in 1×PBS with 0.04% BSA. Liver, lung, and spleen : The fresh tissues were washed by 1×PBS, cut into 2 mm pieces, and incubated with dissociation enzymes (Liver Dissociation Kit, Miltenyi Biotec) at 37℃ for 30–60 min with gentle rotation. Then, the cell suspension was filtered through a 40 µm SmartStrainer, centrifuged at 300 ×g for 10 min at 4℃, and resuspended in 1 mL of 1×PBS. The cell suspension was mixed with Red Cell Lysis Solution at a ratio of 1:3, allowed to stand for 3–5 min, centrifuged at 300 ×g for 5 min at 4℃, and resuspended in 1×PBS with 0.04% BSA. Mammary parenchyma, mammary cistern, mammary duct, esophagus, sublingual gland, intima of uterine horn, intima of corpus uteri, rectum, jejunum, cecum, abomasum, duodenum, ileum, colon, rumen muscle and thymus : The fresh tissues were washed with 1×PBS, cut into 2 mm pieces, and incubated with dissociation enzymes (Collagenase, Type II (Sigma), DNase I (Sigma-Aldrich), and 2% FBS (fetal bovine serum; Gibco)) at 37℃ for 20–45 min with gentle rotation. Following cell dissociation, the cell suspension was filtered through a 70 µm SmartStrainer, centrifuged at 300 ×g for 10 min at 4℃, and resuspended in 1 mL of 1×PBS. The cell suspension was mixed with Red Cell Lysis Solution at a ratio of 1:3, allowed to stand for 3–5 min, centrifuged at 300 ×g for 5 min at 4℃, and resuspended in 1×PBS with 0.04% BSA. Pancreas : The fresh pancreas was washed by 1×PBS, cut into 2 mm pieces, and incubated with dissociation enzymes (Collagenase, Type I (Sigma), DNase I, and 2% FBS) at 37℃ for 20–60 min with gentle rotation. Then, the cell suspension was filtered through a 40 µm SmartStrainer, centrifuged at 300 ×g for 10 min at 4℃, and resuspended in 1 mL of 1×PBS. The cell suspension was mixed with Red Cell Lysis Solution at a ratio of 1:3, allowed to stand for 3–5 min, centrifuged at 300 ×g for 5 min at 4℃, and resuspended in 1×PBS with 0.04% BSA. PBMC : Bovine peripheral blood (3 mL) was gently overlaid onto 3 mL of bovine peripheral blood lymphocyte separation solution (Solarbio) in a 15 mL conical centrifuge tube and centrifuged at 700×g for 30 min at 20°C. After plasma was removed, the buffy coat containing PBMC was collected and transferred into a new 15 mL centrifuge tube containing 10 mL of 1×PBS followed by centrifugation at 300 ×g for 10 min at 4℃. The PBMC was resuspended in 1 mL of 1×PBS. The cell suspension was mixed with Red Cell Lysis Solution at a ratio of 1:3, allowed to stand for 3–8 min, centrifuged at 300 ×g for 5 min at 4℃, and resuspended in 1×PBS with 0.04% BSA. Pituitary gland : The fresh pituitary gland was washed with 1×PBS, cut into 2 mm pieces and incubated with dissociation enzymes (Adult Brain Dissociation Kit, Miltenyi Biotec) at 37℃ for 30–60 min with gentle rotation. Then, the cell suspension was filtered through a 40 µm SmartStrainer and centrifuged at 300×g for 10 min at 4℃. The cells were resuspended in 3.1 mL of 1×PBS and mixed with 900 µL Debris Removal Solution (Miltenyi Biotec). The cell suspension was covered with 4 mL of pre-chilled 1×PBS and centrifuged at 3000 ×g for 10 min at 4℃. The top and second layers of liquid were discarded, and 14 mL of pre-chilled 1×PBS was added. The samples were centrifuged at 1000 ×g for 10 min at 4℃ and resuspended in 1 mL of 1×PBS. The cell suspension was mixed with Red Cell Lysis Solution at a ratio of 1:3, allowed to stand for 3–5 min, centrifuged at 300×g for 5 min at 4℃, and resuspended in 1×PBS with 0.04% BSA. Reticulum, omasum, oviduct, circumvallate papilla, fungal papilla, and rumen papilla : The fresh tissues were washed by 1×PBS, cut into 2 mm pieces, and incubated with dissociation enzymes (Collagenase, Type I, Collagenase, Type II, DNase I, and 2% FBS) at 37℃ for 30–60 min with gentle rotation. Then, the cell suspension was filtered through a 40 µm SmartStrainer, centrifuged at 300 ×g for 10 min at 4℃, and resuspended in 1 mL of 1×PBS. The cell suspension was mixed with Red Cell Lysis Solution at a ratio of 1:3, allowed to stand for 3–5 min, centrifuged at 300 ×g for 5 min at 4℃, and resuspended in 1×PBS with 0.04% BSA. Skin and ovary : The fresh tissues were washed by 1×PBS, cut into 2 mm pieces, and incubated with dissociation enzymes (Collagenase, Type Ⅳ (Sigma), DNase I, and 2% FBS) at 37℃ for 30–60 min with gentle rotation. Then, the cell suspension was filtered through a 40 µm SmartStrainer, centrifuged at 300 ×g for 10 min at 4℃, and resuspended in 1 mL of 1×PBS. The cell suspension was mixed with Red Cell Lysis Solution at a ratio of 1:3, allowed to stand for 3–5 min, centrifuged at 300 ×g for 5 min at 4℃, and resuspended in 1×PBS with 0.04% BSA. Thyroid : The fresh thyroid was washed by 1×PBS, cut into 2 mm pieces, and incubated with dissociation enzymes (Multi Tissue Dissociation Kits 1, Miltenyi Biotec) at 37℃ for 20–60 min with gentle rotation. Then, the cell suspension was filtered through a 70 µm SmartStrainer, centrifuged at 300 ×g for 5 min at 4℃, and resuspended in 1 mL of 1×PBS with 0.04% BSA. The cell suspension was mixed with Red Cell Lysis Solution at a ratio of 1:3, allowed to stand for 3–5 min, centrifuged at 300 ×g for 5 min at 4℃, and resuspended in 1×PBS with 0.04% BSA. Single-nucleus suspension preparation Liver, Bile duct, heart, kidney, cerebral cortex, cerebellum, medulla oblongata, and hypothalamus The single nuclei were isolated using the kit of Isolation of Nuclei for Single Cell RNA Sequencing (10× Genomics). In short, tissues were placed in a 2 mL Eppendorf tube with 1 mL of pre-chilled LB buffer, minced with scissors, and incubated on ice for 1–10 min. The tissue lysate was then filtered with a 40 µm SmartStrainer, centrifuged at 500 ×g for 5 min at 4℃, and resuspended with 300 µL of LB buffer. The nuclei were mixed with 300 µL of RB buffer. Then 600 µL PB1 and 600 µL PB2 buffer were sequentially underlaid from the very bottom of the tube. After centrifugation at 4000 ×g for 20 min at 4℃, the nuclei layer (150–300 µL) was located at the junction of PB1 and PB2 solutions was then aspirated and mixed with 1 mL of RB buffer. The nuclei suspension was then filtered through a 40 µm SmartStrainer, centrifuged at 500 ×g for 5 min at 4℃, and resuspended in 100 µL of EB buffer. Library construction and sequencing by 10× Genomics platform The cell/nucleus suspension (300–600 living cells or nuclei/µL) was loaded onto the Chromium single cell controller (10× Genomics) using Single Cell 3' Library and Gel Bead Kit V3.1 (10× Genomics) and Chromium Single Cell G Chip Kit (10× Genomics) to generate single-cell/nucleus gel beads in emulsion. Briefly, approximately 4,000 cells or nuclei were added to each channel, and the target cells/nuclei recovered were estimated to be about 2,000 cells/nuclei. Captured cells/nuclei were lysed and the released RNA was barcoded through reverse transcription in individual Gel Bead-In Emulsions (GEMs). Reverse transcription was performed on an S1000TM Touch Thermal Cycler (Bio-Rad) at 53°C for 45 min, followed by 85°C for 5 min and at 4°C. The cDNA was generated, amplified, and quality assessed using an Agilent 4200. The libraries were sequenced using an Illumina NovaSeq 6000 sequencer with a sequencing depth of at least 100,000 reads per cell with a pair-end 150 bp (PE150) strategy (CapitalBio Technology Co., Ltd., Beijing, China). Raw sequencing data processing The Bos taurus ARS-UCD1.2 reference assembly 109 in FASTA format and annotated gene model in GTF format were downloaded from the Ensembl database ( ftp://ftp.ensembl.org/pub/release-101 ). The raw scRNA-seq/snRNA-seq data were aligned to the cattle reference genome and subjected to barcode assignment and unique molecular identifier (UMI) counting based on the Cell Ranger 7.0.1 pipeline (10 × Genomics) 110 . For each library, ddqcR 0.1.0 111 was used to remove low-quality cells. It first clustered the cells using standard scRNA-seq analysis preprocessing and clustering steps. In each cluster, cells with n_counts and n_genes values lower than 2 median absolute deviations (MADs) from the median were filtered out. Additionally, cells with a mitochondrial genes percentage lower than 10% were removed. Doublets removal was performed using DoubletFinder 2.0.3 112 with the default parameter. It first averaged the transcriptional profile of randomly chosen cell pairs to create pseudo-doublets and then predicted doublets according to each real cell’s similarity in gene expression to the pseudo-doublets. Cell clusters identification Seurat 4.0.6 113 was used to perform unsupervised clustering. Libraries from the same tissue were merged and underwent normalizing and scaling. Harmony 0.1.1 29 was used to correct four batch effects (sources, methods, platforms, and individuals) with the resetting parameters (Lambda = 1, Theta = 0.5). Variable genes were determined using Seurat’s Find-VariableGenes function with default parameters (selection.method = “vst”, nfeatures = 2000). Clusters were identified via the FindClusters function (resolution = 0.5) implemented in Seurat using the top 30 principal components and subsequently visualized using the RunUMAP function (reduction = “harmony”). Artificial annotation was performed on each cell cluster based on the marker genes reported in the relevant scientific literature. Cell cycle index estimation To obtain additional insights into the dynamic information about cell states, a cell cycle index was computed for each cell type using the CellCycleScoring function within Seurat 113 . The cells were categorized into G0/G1, S, and G2/M phases, denoting distinct cell cycle stages. Cell type preference distribution analysis To characterize the tissue distribution of mammary epithelial cells, odds ratios (ORs) were calculated and used to indicate preferences. Specifically, for each cell subtype i and tissue j, a 2 × 2 matrix was constructed, which contained the number of cells of cell subtype i in tissue j, the number of cells of cell subtype i in other tissues, the number of cells of non-i cell subtype in tissue j, and the number of cells of non-i cell subtype in other tissues. Then Fisher’s exact test was applied to this matrix, thus, OR and the corresponding p -value could be obtained. P -values were adjusted using the Benjamini-Hochberg (BH) method implemented in the R function p.adjust. OR value > 1.5 indicated that cell type i was preferentially distributed in tissue j, and an OR value < 0.5 indicated that cell type i was not preferentially distributed in tissue. Pseudotime trajectory analysis Monocle2 2.26 114 was used to infer the state transition of the cell types/subtypes. The UMI count matrix of the cells was used to create the CellDataSet object and then filter out the genes expressed in fewer than 10 cells. The genes with qvalue < 0.01 were identified as DEGs using the differentialGeneTest function and sorted by qvalue using the setOrderingFilter function. The pseudotime trajectory was constructed by the DDRTree algorithm using the default parameters. The dynamic expression changes of selected marker genes in pseudotime were visualised by the plot_genes_in_pseudotime and plot_pseudotime_heatmap functions. To explore the process of spermatogenesis, the germline cells were extracted from the testis and then re-descended for clustering with “pcs = 20” and resolution set to 0.5. To explore the developmental process of cell types across tissues, cells were extracted and merged from each tissue, batch effects were corrected using Harmony 0.1.1 and then re-descended for clustering with “pcs = 15” and resolution set to 0.1 for epithelial cells, 0.3 for B cells and 0.4 for myeloid cells. In order to shorten the model fitting time, cell subtypes with fewer than 2000 cells in immune cells were completely retained, while subtypes with more than 2000 cells were randomly sampled, and 25% of the cells were retained. To explore the mechanism of monogenic disorders genes in the process of cell differentiation, the uninuclear trophoblast cells and skeletal muscle cells were extracted from the placenta and esophagus, respectively, batch effects were corrected using Harmony 0.1.1 and then re-descended for clustering with “pcs = 30” and resolution set to 0.1. Cellular communication analysis Cellular communication analyses were implemented using the CellChat 1.6.1 R package 38 for each tissue in label-based mode. Default parameters were used, except that min.cells was set to 10, which allows filtering out of cell types with the total number of cells less than 10. All the annotated cell types were classified based on their cell lineage. Interactions between different cell types were then aggregated, and the average intensity was computed to assess the comprehensive dynamics of cellular communication networks. Furthermore, cellular communication patterns within diverse tissues were examined in detail. For comparisons of cellular communication across tissues and species, the cellular communication analysis within specific tissues or species was first performed separately, and then the datasets were merged using mergeCellChat. Finally, the netVisual_diffInteraction function was used to compare and analyze the differences in communication strength. Gene regulatory network analysis Gene regulatory network inference was performed using the Python package pySCENIC 0.12.1 39 with the default parameters. The raw counts, derived from the Seurat object, were based on one-to-one homologous genes between humans and cattle using Homologene 1.4.68.19.3.27. The human TF list ( https://resources.aertslab.org/cistarget/tf_lists/allTFs_hg38.txt ) was used as a reference for identifying co-expression modules through the GRNBoost2 algorithm. Then the regulons were obtained by detecting the genes directly targeted by the TF and removing other genes based on the enrichment of motifs within 10 kb from the target transcription start site (TSS) using cisTarget databases (Homo sapiens - hg38 - refseq_r80 - SCENIC + databases - Gene based (aertslab.org)). Using aucell, the regulon activity score (RAS) was measured as the area under the recovery curve. The activities associated with each cell type were evaluated by calculating the regulon specificity score (RSS) 115 . For cross-species cell-type-specific TFs analysis, the TFs datasets of dairy cattle and humans were separately inferred by pySCENIC. The Jensen Shannon divergence (JSD) was calculated according to the TF expressions, and the TF-specific score was defined as 1-√JSD. The z-score was then calculated to normalize the TF-specific score to predict the basic TF in each cell type by the following formula: \(z score=\frac{{x}_{ij}-{\mu }_{i}}{{\sigma }_{i}}\) ; Where \({x}_{ij}\) was the RSS for TF j in cell type i, \({\mu }_{i}\) was the average RSS of all the TFs in cell type i, \({\sigma }_{i}\) was the standard deviation for RSS of all TFs in cell type i. In total, 500 cells of each cell type were selected for cross-species analyses. The correlation coefficient r and p -value were calculated by corr.test (method = “pearson”, adjust = “fdr”). Transcription factor conservative analysis For each cell type within a given tissue, if a regulon is positive in over 25% of cells, the TF is then considered “positive” in that specific cell type. PhastCons 116 and phyloP 117 conservation scores calculated on multiple sequence alignment of sequences of 100 species of vertebrates were retrieved using UCSC table browser data retrieval tool 118 ( https://hgdownload.soe.ucsc.edu/goldenPath/hg38/ ). TF binding site data in cattle were classified into four groups based on the regulation of cell types, and a Wilcoxon test was conducted to evaluate the significance of differences in TF regulation among these groups. Sex bias classification of tissue analysis The 19 tissues were classified into three categories based on sex bias in cell type composition: non-sex-biased regulation, partial, and sex-biased regulation. Non-sex-biased tissues had no sex-biased cell clusters, partial tissues had less than 50% sex-biased clusters, and tissues with over 50% sex-biased clusters were labeled as sex-biased. The classification was established utilizing the paired Wilcoxon test based on RAS, with subsequent adjustment of p -values using the Bonferroni method for multiple comparisons. Cell type-specific transcription factor network construction For each cell type within a given tissue, a transcription factor (TF) is deemed “positive” in that particular cell type if its regulon is positive in more than 25% of cells, and it also exhibits a z-score value exceeding 0.75. Function enrichment analysis Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were performed using the clusterProfiler 4.0 119 . The GO terms and KEGG pathways of selected genes were enriched in the org.Bt.eg.db and bta databases, using the erichGO and enrichKEGG functions, respectively, with a threshold parameter of “pvalueCutoff = 0.05”. GSVA 1.49.6 120 was used to investigate functional differences between cell subtypes. The bovine KEGG database was downloaded from MsigDB using the msigdbr 7.5.1 ( https://github.com/igordot/msigdbr ), after which the gsva function was used to calculate the pathway scores in the pathway for each subtype-specific gene set. Cell type diversity analysis The Shannon entropy 121 was calculated to evaluate cell type diversity in each tissue according to the formula \(-{\sum }_{\text{i}}({\text{p}}_{\text{i}}\times {\text{l}\text{o}\text{g}}_{2}({\text{p}}_{\text{i}}\left)\right)\) , where \({\text{p}}_{\text{i}}\) is the proportion of cell type in cell class i for each tissue. They were then plotted in R 4.2.0 using ggplot2 3.4.1 and complexHeatmap 2.15.4 122 . Cell type conservation analysis MetaNeighbor 1.18 123 was used to provide a measure of the replicability of cell types within and across species. The highly variable genes were identified using the variableGenes function and the correlations between cell types were determined based on AUROC values using MetaNeighborUS analysis. In the cross-species analysis, 500 cells of each cell type were selected. Cell function scoring To evaluate the extracellular and intracellular antigen presentation abilities of myeloid cells and B cells, as well as macrophage subpopulations of the M1 and M2 phenotypes, the AddModuleScore function in the Seurat package was used to calculate the antigen presentation score (APS) and macrophage_type score. Using the “special = Bos taurus, category = C5, subcategory = GO: BP” parameter of the msigdbr package ( https://github.com/igordot/msigdbr ) to obtain gene set ( Supplementary Table 4 − 2 ) from “GOBP_ANTIGEN_PROCESSING_AND_PRESENTATION_OF_EXOGENOUS_PEPTIDE_ANTIGEN_VIA_MHC_CLASS_I”, “GOBP_ANTIGEN_PROCESSING_AND_PRESENTATION_OF_PEPTIDE_ANTIGEN_VIA_MHC_CLASS_I”, “GOBP_ANTIGEN_PROCESSING_AND_PRESENTATION_OF_EXOGENOUS_PEPTIDE_ANTIGEN_VIA_MHC_CLASS_II”, “GOBP_ANTIGEN_PROCESSING_AND_PRESENTATION_OF_PEPTIDE_OR_POLYSACCHARIDE_ANTIGEN_VIA_MHC_CLASS_II” in the msigdb database ( https://www.gsea-msigdb.org/gsea/msigdb ). The classically activated M1 macrophages gene sets are SOCS1 , NOS2 , TNF , CXCL9 , CXCL10, CXCL11 , CD86 , IL1A , IL1B , IL6 , CCL5 , IRF5 , IRF1 , and CCR7 , while the selectively activated M2 macrophages gene sets are IL4R , CCL4 , CCL18 , CCL22 , MARCO , VEGFA , CTSA , CTSB , TGFB1 , MMP9 , CLEC7A , MSR1 , IRF4 , CD163 , TGM2 , and MRC1 . Data were compared by the two-tailed Student’s t -test (**** p 0.05). High-dimensional weighted correlation network analysis The hdWGCNA 0.2.24 124 was used to perform high-dimensional weighted gene co-expression network analysis based on single-cell data. First, we input the genes expressed in at least 5% of the cells and used the MetacellsByGroups function to construct the metacell gene expression matrix. Then, the TestSoftPowers function was used to determine the soft power. The ConstructNetwork function was used to build the co-expression network. All analyses were conducted according to standard procedures ( https://smorabit.github.io/hdWGCNA/articles/hdWGCNA.html ). Cattle monogenic disorder genes collection A total of 634 monogenic disorders in cattle were assembled from the Online Mendelian Inheritance in Animals (OMIA) database ( https://www.omia.org/home ). Candidate genes related to 145 disorders were identified. To facilitate systematic analysis, these disorders were classified into 10 groups based on their phenotypic manifestations, encompassing conditions related to blood and immune systems, connective tissue, embryonic development, embryonic lethality, metabolic processes, muscle function, neural disorders, reproductive issues, skin-related disorders, and tissue-specific conditions ( Supplementary Table 6 − 1 ). In order to ensure the robustness of the statistical analyses, a minimum threshold of more than five candidate genes was imposed for each disorder group. Cell-type specific gene identification Cell-type specific genes were identified using the z-score factor. Initially, the count matrix underwent a transformation into the Counts Per Million (CPM) values, and the average was computed within each cell type to create a pseudo-bulk expression matrix 17 . Subsequently, tspex 0.6.2 125 was utilized to calculate z-score values under the log10-transformed CPM matrix. Genes exhibiting a z-score value > 0.75 in each cell type were designated as cell-type-specific genes and cell-type-specific genes in a given cell lineage were defined as cell lineage-specific genes. Enrichment analysis between cell types and monogenic disorders The correlation between cell-type-specific genes and monogenic disorder genes was assessed through Chi-square and Fisher’s tests, utilizing a 2×2 matrix consisting of intersected genes, disorder genes, cell-type-specific genes, and all genes. The Chi-square test was employed when expected frequencies for all elements exceeded 5; otherwise, Fisher’s exact test was used. Subsequently, the false discovery rate (FDR) values were computed using the Bonferroni method to address multiple comparisons. Additionally, ORs were also calculated to validate the accuracy of significant results based on the same 2×2 matrix. Gene coding DNA sequence (CDS) region alignment between humans and cattle DNA sequences of the gene CDS region were downloaded from the NCBI-Genbank ( https://www.ncbi.nlm.nih.gov/genbank/ ). Sequence alignment was performed based on the NCBI-Blast method ( https://blast.ncbi.nlm.nih.gov/Blast.cgi ). Genome-wide association study (GWAS) analysis The SNP BeadChip and phenotypic data of 16,188 Chinese Holstein cows were assembled, including 9,045 bovine 150K BeadChip, 1,505 bovine 80K BeadChip, and 5,638 bovine 50K BeadChip, from the Dairy Association of China. The phenotypic data included 55 traits, spanning milk production, sperm, coat color, lgG, body conformation, and health. All SNP BeadChip data were mapped to the bovine reference genome (ARS-UCD1.2) and 132,961 SNPs were obtained after imputing to the 150K level using Beagle 5.1 126 . They were then imputed again to genome-scale sequence level using a high-quality sequencing imputation panel of 28,166,177 SNPs based on 3,530 cattle. After filtering out low-quality SNPs with “Dosage R-Squared (DR2) < 0.9, Minor Allele Frequency (MAF) < 0.05 and Hardy-Weinberg Equilibrium (HWE) test result P < 0.0001”, a total of 8,535,460 SNPs were obtained. Finally, GWAS analysis was performed using GCTA 1.94.0 127 with the “--fastGWA-mlm” option, and the results were visualized using CMplot 4.4.1 128 . Enrichment analysis between cell types and complex traits Scpagwas 1.3.0 84 was utilized to perform the enrichment analysis between cell types and complex traits. It employs a polygenic regression model to prioritize a set of trait-relevant genes and uncover trait-relevant cell subpopulations by incorporating pathway activity transformed scRNA-seq data with GWAS summary data. To enhance the comprehensiveness of our results, 319 human KEGG pathways downloaded from the KEGG database ( https://www.genome.jp/kegg ) were used after eliminating duplicates and converting homologous genes. The Boot_evaluate function was employed to identify the significant trait-relevant relevant cell types and calculate trait-relevant scores. The scGet_PCC function was used to prioritize the top trait-relevant genes by ranking the Pearson correlation coefficient (PCC). Genes with the top 50 PCC values were defined as trait-relevant genes in each cell type. In addition, the scPagwas_perform_score function was applied to perform pathway activity analysis and define the significance of active pathways in each cell type based on the singular value decomposition (SVD) method. Enrichment analysis between trait-relevant genes and active pathway genes was performed based on Fisher’s test and Chi-square test within each cell type. Linkage disequilibrium score regression (LDSC) LDSC 129 was to detect whether the heritability of a phenotype is enriched around highly specifically expressed genes in a given tissue or cell type. All SNPs associated with the trait were obtained from publicly available data ( Supplementary Table 8 − 3 ). The comparison of tissue contributions involved the selection of the top 200 DEGs (log 2 FC ≥ 1.5 and FDR ≤ 0.05), which were sorted by FDR from least to most for each tissue as a category. All DEG categories for tissues were collectively inputted to run LDSC for traits. Subsequently, the p -values were computed using the BH method to account for multiple comparisons. Similarly, the top 200 DEGs of each cell type cluster meeting the same criteria were selected for analysis, and the results were visualized using ggplot2 3.4.1. Statistical and reproducibility No statistical method was used to predetermine the sample size. The details of data exclusions for each specific analysis are available in the Methods section. The experiments were not randomized, as all the datasets are publicly available from observational studies. The investigators were not blinded to allocation during experiments and outcome assessment, as the data were not from controlled randomized studies. Declarations Data and code availability All single-cell and single-nucleus RNA sequencing data newly generated in this study are available for download under accession number PRJNA1119173 from SRA ( https://www.ncbi.nlm.nih.gov/sra/ ) via the provided reviewer link ( https://dataview.ncbi.nlm.nih.gov/object/PRJNA1119173?reviewer=po13jm18jl62b78mi9k7r1mju7/ ). The processed datasets and expression profiles of annotated cell types are available via the CattleCA web portal: https://ngdc.cncb.ac.cn/cattleca/ . All the computational codes are freely available via https://github.com/FarmGTEx/CattleCellAtlas_pipeline_V0/ . Acknowledgements This work was financially supported by the National Key R&D Program of China (2021YFF1000700, 2022YFF1000103); the National Natural Science Foundation of China (32372836); STI 2030-Major Projects (2023ZD04069); and the Program for Changjiang Scholar and Innovation Research Team in University (IRT_15R62). Lingzhao Fang was supported from Agriculture and Food Research Initiative Competitive grants nos. 2022-67015-36215 (H.Z.) from the USDA National Institute of Food and Agriculture. Jayne C. Hope was funded by the Biotechnology and Biological Sciences Research Council through Institute Strategic Programme Funding (grant number BBS/E/RL/230002B). J.F.O. is supported by the Science Foundation Ireland (SFI) Centre for Research Training in Genomics Data Science (grant no. 18/CRT/6214). Bingjie Li acknowledges funding from the UK Biotechnology and Biological Sciences Research Council (BBSRC) with grant BB/X009505/1. We acknowledge the support of the High-Performance Computing Platform of China Agricultural University (Beijing) and the Xihe High-Performance Computing Platform of the National Research Facility for Phenotypic and Genotypic Analysis of Model Animals (Beijing). Author contributions All authors made substantial contributions to the conception or design of the study; the acquisition, analysis or interpretation of data; or drafting or revising the paper. D. Sun, L. Fang, B. Han, H. Sun, Y. Jiang, and G. E. Liu conceived and designed the project. D. Sun, Y. Jiang, H. Sun, Z. Ma, and L. Liu provided samples and data. H. Li, B. Han, S. Zhu, and T. Shi performed bioinformatic analyses of single cell/nucleus RNA-seq data. Q. Zhang, A. Chen, Y. Song, W. Ye, A. Du, Y. Fu, M. Jia, and T. Shi annotated the cell types manually. Q. Zhang and B. Han conducted the intra-tissue cellular heterogeneity. W. Zheng and B. Han conducted the inter-tissue cellular heterogeneity. H. Li and W. Zheng conducted integrative analysis with genetic variants. A. Chen and B. Han conducted comparative analysis between cattle and humans. Y. Hou, Z. Zhang, D. Zou, and Z. Yuan built the CattleCA web portal. D. Sun, L. Fang, and B. Han contributed to the data and computational resources. D. Sun, L. Fang, G. E. Liu, Y. Hou, F. Wang, H. Sun, Y. Jiang, W. Liu, W. Tuo, and J. C. Hope contributed to the critical interpretation of analytical results before and during manuscript preparation. B. Han, H. Li, Q. Zhang, W. Zheng, and A. Chen drafted the manuscript. L. Fang, D. Sun, G. E. Liu, Y. Hou, F. Wang, H. Sun, Y. Jiang, W. Liu, W. Tuo, J. C. Hope, D. E. MacHugh, J. F. O. Grady, O. Madsen, G. Sahana, Y. Luo, L. Lin, C. Li, Z. Cai, B. Li, and Z. Zhang revised the manuscript. All authors read, edited and approved the final manuscript. Competing interests The authors declare no competing interests. References Bruford, M.W., Bradley, D.G. & Luikart, G. DNA markers reveal the complexity of livestock domestication. Nat Rev Genet 4, 900–10 (2003). Ajmone-Marsan, P., Garcia, J.F. & Lenstra, J.A. On the origin of cattle: How aurochs became cattle and colonized the world. 19, 148–157 (2010). Pitt, D. et al. Domestication of cattle: Two or three events? Evol Appl 12, 123–136 (2019). Ritter, C., Beaver, A. & von Keyserlingk, M.A.G. The complex relationship between welfare and reproduction in cattle. Reproduction in Domestic Animals 54, 29–37 (2019). Laborde, D., Mamun, A., Martin, W., Piñeiro, V. & Vos, R. Agricultural subsidies and global greenhouse gas emissions. Nature Communications 12(2021). Xu, X.M. et al. Global greenhouse gas emissions from animal-based foods are twice those of plant-based foods. Nature Food 2, 724-+ (2021). Hawkins, J.W. et al. High-yield dairy cattle breeds improve farmer incomes, curtail greenhouse gas emissions and reduce dairy import dependency in Tanzania. Nature Food 3, 957-+ (2022). Hu, Z.-L., Park, C.A. & Reecy, J.M. Bringing the Animal QTLdb and CorrDB into the future: meeting new challenges and providing updated services. Nucleic Acids Research 50, D956-D961 (2021). Freebern, E. et al. GWAS and fine-mapping of livability and six disease traits in Holstein cattle. BMC Genomics 21, 41 (2020). Hosseinzadeh, S., Rafat, S.A., Javanmard, A. & Fang, L. Identification of candidate genes associated with milk production and mastitis based on transcriptome-wide association study. Anim Genet 55, 430–439 (2024). Jiang, J. et al. Functional annotation and Bayesian fine-mapping reveals candidate genes for important agronomic traits in Holstein bulls. Commun Biol 2, 212 (2019). Jiang, L. et al. Genome wide association studies for milk production traits in Chinese Holstein population. PLoS One 5, e13661 (2010). Liu, S. et al. A multi-tissue atlas of regulatory variants in cattle. Nat Genet 54, 1438–1447 (2022). Han, X. et al. Construction of a human cell landscape at single-cell level. Nature 581, 303–309 (2020). Tabula Sapiens, C. et al. The Tabula Sapiens: A multiple-organ, single-cell transcriptomic atlas of humans. Science 376, eabl4896 (2022). Han, X. et al. Mapping the Mouse Cell Atlas by Microwell-Seq. Cell 173, 1307 (2018). Li, H. et al. Fly Cell Atlas: A single-nucleus transcriptomic atlas of the adult fruit fly. Science 375, eabk2432 (2022). Wagner, D.E. et al. Single-cell mapping of gene expression landscapes and lineage in the zebrafish embryo. Science 360, 981–987 (2018). Ye, F. et al. Construction of the axolotl cell landscape using combinatorial hybridization sequencing at single-cell resolution. Nat Commun 13, 4228 (2022). Wang, F. et al. Endothelial cell heterogeneity and microglia regulons revealed by a pig cell landscape at single-cell level. Nat Commun 13, 3620 (2022). Gao, Y. et al. Single-cell transcriptomic analyses of dairy cattle ruminal epithelial cells during weaning. Genomics 113, 2045–2055 (2021). Wu, J.J. et al. Microbiota-host crosstalk in the newborn and adult rumen at single-cell resolution. BMC Biol 20, 280 (2022). Xue, M.Y. et al. Investigation of fiber utilization in the rumen of dairy cows based on metagenome-assembled genomes and single-cell RNA sequencing. Microbiome 10, 11 (2022). Gao, Y. et al. Single-cell transcriptomic and chromatin accessibility analyses of dairy cattle peripheral blood mononuclear cells and their responses to lipopolysaccharide. BMC Genomics 23, 338 (2022). Gu, F. et al. Gut microbiome is linked to functions of peripheral immune cells in transition cows during excessive lipolysis. Microbiome 11, 40 (2023). Wang, L. et al. A single-cell atlas of bovine skeletal muscle reveals mechanisms regulating intramuscular adipogenesis and fibrogenesis. J Cachexia Sarcopenia Muscle 14, 2152–2167 (2023). Wu, J.J. et al. Cross-tissue single-cell transcriptomic landscape reveals the key cell subtypes and their potential roles in the nutrient absorption and metabolism in dairy cattle. J Adv Res 37, 1–18 (2022). Wu, J.J. et al. Age- and Microbiota-Dependent Cell Stemness Plasticity Revealed by Cattle Cell Landscape. Research (Wash D C) 6, 0025 (2023). Korsunsky, I. et al. Fast, sensitive and accurate integration of single-cell data with Harmony. Nat Methods 16, 1289–1296 (2019). Strandberg, Y. et al. Lipopolysaccharide and lipoteichoic acid induce different innate immune responses in bovine mammary epithelial cells. Cytokine 31, 72–86 (2005). Morato, A., Martignani, E., Miretti, S., Baratta, M. & Accornero, P. External and internal EGFR-activating signals drive mammary epithelial cells proliferation and viability. Molecular and Cellular Endocrinology 520(2021). Mipam, T. et al. Single-cell transcriptome analysis and in vitro differentiation of testicular cells reveal novel insights into male sterility of the interspecific hybrid cattle-yak. Bmc Genomics 24(2023). Guo, J.T. et al. The adult human testis transcriptional cell atlas. Cell Research 28, 1141–1157 (2018). Wu, S., Hu, Y.C., Liu, H.F. & Shi, Y. Loss of YY1 Impacts the Heterochromatic State and Meiotic Double-Strand Breaks during Mouse Spermatogenesis. Molecular and Cellular Biology 29, 6245–6256 (2009). Wawrzik, M., Spiess, A.N., Herrmann, R., Buiting, K. & Horsthemke, B. Expression of SNURF–SNRPN upstream transcripts and epigenetic regulatory genes during human spermatogenesis. European Journal of Human Genetics 17, 1463–1470 (2009). Fimia, G.M., De Cesare, D. & Sassone-Corsi, P. CBP-independent activation of CREM and CREB by the LIM-only protein ACT. Nature 398, 165–169 (1999). Bechtel, T.J., Reyes-Robles, T., Fadeyi, O.O. & Oslund, R.C. Strategies for monitoring cell-cell interactions. Nat Chem Biol 17, 641–652 (2021). Jin, S.Q. et al. Inference and analysis of cell-cell communication using CellChat. Nature Communications 12(2021). Aibar, S. et al. SCENIC: single-cell regulatory network inference and clustering. Nature Methods 14, 1083-+ (2017). Fischer, C. et al. AGMO Inhibitor Reduces 3T3-L1 Adipogenesis. Cells 10(2021). Helsley, R.N. et al. Ketohexokinase-C regulates global protein acetylation to decrease carnitine palmitoyltransferase 1a-mediated fatty acid oxidation. Journal of Hepatology 79(2023). Wang, S.Y. et al. Loss of hepatic FTCD promotes lipid accumulation and hepatocarcinogenesis by upregulating PPARc and SREBP2. Jhep Reports 5(2023). Chen, T. et al. Hippocalcin-Like 1 blunts liver lipid metabolism to suppress tumorigenesis via directly targeting RUVBL1-mTOR signaling. Theranostics 12, 7450–7464 (2022). McLatchie, L.M. et al. RAMPs regulate the transport and ligand specificity of the calcitonin-receptor-like receptor. Nature 393, 333–339 (1998). Stutchfield, B.M. & Forbes, S.J. Liver sinusoidal endothelial cells in disease–and for therapy? J Hepatol 58, 178–80 (2013). Hammoutene, A. & Rautou, P.E. Role of liver sinusoidal endothelial cells in non-alcoholic fatty liver disease. J Hepatol 70, 1278–1291 (2019). Vyas, J.M., Van der Veen, A.G. & Ploegh, H.L. The known unknowns of antigen processing and presentation. Nature Reviews Immunology 8, 607–618 (2008). Pishesha, N., Harmand, T.J. & Ploegh, H.L. A guide to antigen processing and presentation. Nature Reviews Immunology 22, 751–764 (2022). Mass, E., Nimmerjahn, F., Kierdorf, K. & Schlitzer, A. Tissue-specific macrophages: how they develop and choreograph tissue biology. Nature Reviews Immunology (2023). Hashimoto, D. et al. Tissue-Resident Macrophages Self-Maintain Locally throughout Adult Life with Minimal Contribution from Circulating Monocytes. Immunity 38, 792–804 (2013). Davies, L.C., Jenkins, S.J., Allen, J.E. & Taylor, P.R. Tissue-resident macrophages. Nature Immunology 14, 986–995 (2013). Schuijs, M.J., Hammad, H. & Lambrecht, B.N. Professional and 'Amateur' Antigen-Presenting Cells In Type 2 Immunity. Trends in Immunology 40, 22–34 (2019). Roche, P.A. & Furuta, K. The ins and outs of MHC class II-mediated antigen processing and presentation. Nature Reviews Immunology 15, 203–216 (2015). Green, M.R. et al. Signatures of murine B-cell development implicate Yy1 as a regulator of the germinal center-specific program. Proc Natl Acad Sci U S A 108, 2873–8 (2011). Huang, C., Geng, H., Boss, I., Wang, L. & Melnick, A. Cooperative transcriptional repression by BCL6 and BACH2 in germinal center B-cell differentiation. Blood 123, 1012–20 (2014). Cousu, C. et al. Germinal center output is sustained by HELLS-dependent DNA-methylation-maintenance in B cells. Nat Commun 14, 5695 (2023). Calame, K.L., Lin, K.I. & Tunyaplin, C. Regulatory mechanisms that determine the development and function of plasma cells. Annu Rev Immunol 21, 205–30 (2003). He, S. et al. Single-cell transcriptome profiling of an adult human cell atlas of 15 major organs. Genome Biol 21, 294 (2020). Engevik, A.C., Kaji, I. & Goldenring, J.R. The Physiology of the Gastric Parietal Cell. Physiol Rev 100, 573–602 (2020). Groschwitz, K.R. & Hogan, S.P. Intestinal barrier function: molecular regulation and disease pathogenesis. J Allergy Clin Immunol 124, 3–20; quiz 21 – 2 (2009). Chen, L. et al. Large-scale ruminant genome sequencing provides insights into their evolution and distinct traits. Science 364(2019). Soubeyran, P. et al. Cdx1 promotes differentiation in a rat intestinal epithelial cell line. Gastroenterology 117, 1326–38 (1999). Silberg, D.G., Swain, G.P., Suh, E.R. & Traber, P.G. Cdx1 and cdx2 expression during intestinal development. Gastroenterology 119, 961 – 71 (2000). Gustafsson, J.K. & Johansson, M.E.V. The role of goblet cells and mucus in intestinal homeostasis. Nat Rev Gastroenterol Hepatol 19, 785–803 (2022). Farr, L. et al. CD74 Signaling Links Inflammation to Intestinal Epithelial Cell Regeneration and Promotes Mucosal Healing. Cell Mol Gastroenterol Hepatol 10, 101–112 (2020). Lees, C.W., Barrett, J.C., Parkes, M. & Satsangi, J. New IBD genetics: common pathways with other diseases. Gut 60, 1739–53 (2011). Muise, A.M. et al. Polymorphisms in E-cadherin (CDH1) result in a mis-localised cytoplasmic protein that is associated with Crohn's disease. Gut 58, 1121–7 (2009). Nicholas, F.W. Online Mendelian Inheritance in Animals (OMIA): a record of advances in animal genetics, freely available on the Internet for 25 years. Anim Genet 52, 3–9 (2021). Bolling, M.C. et al. Lethal acantholytic epidermolysis bullosa due to a novel homozygous deletion in DSP: expanding the phenotype and implications for desmoplakin function in skin and heart. Br J Dermatol 162, 1388–94 (2010). He, H.J., Bing, H. & Liu, G. TSR2 Induces laryngeal cancer cell apoptosis through inhibiting NF-kappaB signaling pathway. Laryngoscope 128, E130-E134 (2018). Akiyama, M. The roles of ABCA12 in keratinocyte differentiation and lipid barrier formation in the epidermis. Dermatoendocrinol 3, 107 – 12 (2011). Betz, R.C. et al. Loss-of-function mutations in the keratin 5 gene lead to Dowling-Degos disease. Am J Hum Genet 78, 510–9 (2006). Fukuda, M., Kuroda, T.S. & Mikoshiba, K. Slac2-a/melanophilin, the missing link between Rab27 and myosin Va: implications of a tripartite protein complex for melanosome transport. J Biol Chem 277, 12432–6 (2002). Graf, J., Hodgson, R. & van Daal, A. Single nucleotide polymorphisms in the MATP gene are associated with normal human pigmentation variation. Hum Mutat 25, 278–84 (2005). Aumailley, M. et al. A simplified laminin nomenclature. Matrix Biol 24, 326–32 (2005). Kiritsi, D., Has, C. & Bruckner-Tuderman, L. Laminin 332 in junctional epidermolysis bullosa. Cell Adh Migr 7, 135–41 (2013). Liu, M., Yin, Y., Yu, H. & Zhou, R. Laminins Regulate Placentation and Pre-eclampsia: Focus on Trophoblasts and Endothelial Cells. Front Cell Dev Biol 8, 754 (2020). Uitto, J. & Richard, G. Progress in epidermolysis bullosa: genetic classification and clinical implications. Am J Med Genet C Semin Med Genet 131C, 61–74 (2004). Wiedemar, N., Riedi, A.K., Jagannathan, V., Drögemüller, C. & Meylan, M. Genetic Abnormalities in a Calf with Congenital Increased Muscular Tonus. Journal of Veterinary Internal Medicine 29, 1418–1421 (2015). Shashi, V. et al. Heterozygous variants in MYBPC1 are associated with an expanded neuromuscular phenotype beyond arthrogryposis. Human Mutation 40, 1115–1126 (2019). Weber, F.E., Vaughan, K.T., Reinach, F.C. & Fischman, D.A. Complete sequence of human fast-type and slow-type muscle myosin-binding-protein C (MyBP-C). Differential expression, conserved domain structure and chromosome assignment. Eur J Biochem 216, 661–9 (1993). Simpson, M.A. et al. A mutation in NFkappaB interacting protein 1 causes cardiomyopathy and woolly haircoat syndrome of Poll Hereford cattle. Anim Genet 40, 42–6 (2009). Ganguly, J., Kulshreshtha, D., Almotiri, M. & Jog, M. Muscle Tone Physiology and Abnormalities. Toxins (Basel) 13(2021). Ma, Y. et al. Polygenic regression uncovers trait-relevant cellular contexts through pathway activation transformation of single-cell RNA sequencing data. Cell Genom 3, 100383 (2023). Fang, L.Z. et al. Comprehensive analyses of 723 transcriptomes enhance genetic and biological interpretations for complex traits in cattle. Genome Research 30, 790–801 (2020). Li, M. et al. Gut-brain circuits for fat preference. Nature 610, 722–730 (2022). Linehan, V., Fang, L.Z. & Hirasawa, M. Short-term high-fat diet primes excitatory synapses for long-term depression in orexin neurons. J Physiol 596, 305–316 (2018). Wallace, C.W. & Fordahl, S.C. Obesity and dietary fat influence dopamine neurotransmission: exploring the convergence of metabolic state, physiological stress, and inflammation on dopaminergic control of food intake. Nutr Res Rev 35, 236–251 (2022). Sacchetti, M., Saltin, B., Osada, T. & van Hall, G. Intramuscular fatty acid metabolism in contracting and non-contracting human skeletal muscle. J Physiol 540, 387–95 (2002). Watt, M.J. & Hoy, A.J. Lipid metabolism in skeletal muscle: generation of adaptive and maladaptive intracellular signals for cellular function. Am J Physiol Endocrinol Metab 302, E1315-28 (2012). Twigger, A.J. et al. Transcriptional changes in the mammary gland during lactation revealed by single cell sequencing of cells from human milk. Nat Commun 13, 562 (2022). Neto, F.T., Bach, P.V., Najari, B.B., Li, P.S. & Goldstein, M. Spermatogenesis in humans and its affecting factors. Semin Cell Dev Biol 59, 10–26 (2016). Duan, Y.G. et al. Dendritic cells in semen of infertile men: association with sperm quality and inflammatory status of the epididymis. Fertil Steril 101, 70–77 e3 (2014). Chen, L. et al. A reinforcing HNF4–SMAD4 feed-forward module stabilizes enterocyte identity. 51, 777–785 (2019). Gao, N., White, P. & Kaestner, K.H. Establishment of intestinal identity and epithelial-mesenchymal signaling by Cdx2. Developmental cell 16, 588–599 (2009). Tugores, A. et al. The epithelium-specific ETS protein EHF/ESE-3 is a context-dependent transcriptional repressor downstream of MAPK signaling cascades. Journal of Biological Chemistry 276, 20397–20406 (2001). Flandez, M., Guilmeau, S., Blache, P. & Augenlicht, L. KLF4 regulation in intestinal epithelial cell maturation. Experimental cell research 314, 3712–3723 (2008). Rouco, R. et al. Cell-specific alterations in Pitx1 regulatory landscape activation caused by the loss of a single enhancer. 12, 7235 (2021). Naser, S.A., Sagramsingh, S.R., Naser, A.S. & Thanigachalam, S.J.W.J.o.G.W. Mycobacterium avium subspecies paratuberculosis causes Crohn's disease in some inflammatory bowel disease patients. 20, 7403 (2014). Bendele, A. et al. Animal models of arthritis: relevance to human disease. 27, 134–142 (1999). Hamel, A.R. et al. Integrating genetic regulation and single-cell expression with GWAS prioritizes causal genes and cell types for glaucoma. Nature Communications 15(2024). McDaniel, C.J., Cardwell, D.M., Moeller, R.B. & Gray, G.C. Humans and Cattle: A Review of Bovine Zoonoses. Vector-Borne and Zoonotic Diseases 14, 1–19 (2014). Saied, A.A., Metwally, A.A., Mohamed, H.M.A. & Haridy, M.A.M. The contribution of bovines to human health against viral infections. Environmental Science and Pollution Research 28, 46999–47023 (2021). Hamernik, D.L. Farm animals are important biomedical models. Animal Frontiers 9, 3–5 (2019). Zeng, H.K. What is a cell type and how to define it? Cell 185, 2739–2755 (2022). Soto, D.A. & Ross, P.J. Similarities between bovine and human germline development revealed by single-cell RNA sequencing. Reproduction 161, 239–253 (2021). Davenport, K.M. et al. Single-nuclei RNA sequencing (snRNA-seq) uncovers trophoblast cell types and lineages in the mature bovine placenta. Proc Natl Acad Sci U S A 120, e2221526120 (2023). Panebianco, C.J., Dave, A., Charytonowicz, D., Sebra, R. & Iatridis, J.C. Single-cell RNA-sequencing atlas of bovine caudal intervertebral discs: Discovery of heterogeneous cell populations with distinct roles in homeostasis. FASEB J 35, e21919 (2021). Rosen, B.D. et al. De novo assembly of the cattle reference genome with single-molecule sequencing. Gigascience 9(2020). Zheng, G.X. et al. Massively parallel digital transcriptional profiling of single cells. Nat Commun 8, 14049 (2017). Subramanian, A., Alperovich, M., Yang, Y. & Li, B. Biology-inspired data-driven quality control for scientific discovery in single-cell transcriptomics. Genome Biol 23, 267 (2022). McGinnis, C.S., Murrow, L.M. & Gartner, Z.J. DoubletFinder: Doublet Detection in Single-Cell RNA Sequencing Data Using Artificial Nearest Neighbors. Cell Syst 8, 329–337 e4 (2019). Hao, Y. et al. Integrated analysis of multimodal single-cell data. Cell 184, 3573–3587 e29 (2021). Trapnell, C. et al. The dynamics and regulators of cell fate decisions are revealed by pseudotemporal ordering of single cells. Nature Biotechnology 32, 381–386 (2014). Suo, S.B. et al. Revealing the Critical Regulators of Cell Identity in the Mouse Cell Atlas. Cell Reports 25, 1436-+ (2018). Siepel, A. et al. Evolutionarily conserved elements in vertebrate, insect, worm, and yeast genomes. Genome Research 15, 1034–1050 (2005). Pollard, K.S., Hubisz, M.J., Rosenbloom, K.R. & Siepel, A. Detection of nonneutral substitution rates on mammalian phylogenies. Genome Research 20, 110–121 (2010). Karolchik, D. et al. The UCSC Table Browser data retrieval tool. Nucleic Acids Research 32, D493-D496 (2004). Wu, T. et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation (Camb) 2, 100141 (2021). Hänzelmann, S., Castelo, R. & Guinney, J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics 14, 7 (2013). Eraslan, G. et al. Single-nucleus cross-tissue molecular reference maps toward understanding disease gene function. Science 376, eabl4290 (2022). Gu, Z., Eils, R. & Schlesner, M. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics 32, 2847–9 (2016). Crow, M., Paul, A., Ballouz, S., Huang, Z.J. & Gillis, J. Characterizing the replicability of cell types defined by single cell RNA-sequencing data using MetaNeighbor. Nature Communications 9(2018). Morabito, S. et al. Single-nucleus chromatin accessibility and transcriptomic characterization of Alzheimer's disease. Nat Genet 53, 1143–1155 (2021). Camargo, A.P., Vasconcelos, A.A., Fiamenghi, M.B., Pereira, G.A.G. & Carazzolle, M.F. tspex: a tissue-specificity calculator for gene expression data. (2020). Browning, B.L., Zhou, Y. & Browning, S.R. A One-Penny Imputed Genome from Next-Generation Reference Panels. Am J Hum Genet 103, 338–348 (2018). Jiang, L., Zheng, Z., Fang, H. & Yang, J. A generalized linear mixed model association tool for biobank-scale data. Nat Genet 53, 1616–1621 (2021). Yin, L. et al. rMVP: A Memory-efficient, Visualization-enhanced, and Parallel-accelerated Tool for Genome-wide Association Study. Genomics Proteomics Bioinformatics 19, 619–628 (2021). Bulik-Sullivan, B.K. et al. LD Score regression distinguishes confounding from polygenicity in genome-wide association studies. Nature genetics 47, 291–295 (2015). Additional Declarations There is NO Competing Interest. Supplementary Files CattleCellAtalsSupplementaryinformation.docx SupplementaryTable11.xlsx SupplementaryTable12.xlsx SupplementaryTable13.xlsx SupplementaryTable14.xlsx SupplementaryTable21.xlsx SupplementaryTable31.xlsx SupplementaryTable41.xlsx SupplementaryTable42.xls SupplementaryTable51.xlsx SupplementaryTable52.xlsx SupplementaryTable61.xlsx SupplementaryTable62.xlsx SupplementaryTable63.xlsx SupplementaryTable71.xlsx SupplementaryTable72.xlsx SupplementaryTable81.xlsx SupplementaryTable82.xlsx SupplementaryTable83.xlsx Cite Share Download PDF Status: Published Journal Publication published 05 Sep, 2025 Read the published version in Nature Genetics → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4631710","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":321685955,"identity":"444d73cf-b5c9-4904-9383-7f221b725323","order_by":0,"name":"Lingzhao Fang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCUlEQVRIiWNgGAWjYBACPmYgwdgAJJgPAIkKmDhbAk4tbHAtYEVniNHCgKyFsY0YLey8h1/+3GEnB7Tv2cOv86zt5jswP5NgKEvD4zC+NAvJM8nGDGxs5say29KTNx5gM5NgOJeDRwuPmYFhG3Nig3yDmbTktsPJhg0MZhKMbRX4tSS21Sc2sLF/k5acA9LC/o2QFuMHB9sOA7XwmEl+bDhsJ8/AA7IFv8MYG9uOG7Ox8ZRJMxxLTzBg5im2SDiH2/v8/GeMP/5sq5bjZ2PfJvmjxtpevr19440PZck4tYAskgCTQMzMw8CcuOEwkJWATwNQ4QcYi/EHA7O9fAN+5aNgFIyCUTDyAAA4TEdeX6HzoAAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0003-1103-3679","institution":"Aarhus University","correspondingAuthor":true,"prefix":"","firstName":"Lingzhao","middleName":"","lastName":"Fang","suffix":""},{"id":321685956,"identity":"ad2da654-5e1a-4c8f-be70-5f2a8748e9dd","order_by":1,"name":"Bo Han","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Bo","middleName":"","lastName":"Han","suffix":""},{"id":321685957,"identity":"04cdefcf-af7f-4ba6-8b2e-5cbb71d42bf9","order_by":2,"name":"Houcheng Li","email":"","orcid":"","institution":"Aarhus University","correspondingAuthor":false,"prefix":"","firstName":"Houcheng","middleName":"","lastName":"Li","suffix":""},{"id":321685958,"identity":"98fdf0fe-7441-4381-9adb-25b62bd62bed","order_by":3,"name":"Qi Zhang","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Qi","middleName":"","lastName":"Zhang","suffix":""},{"id":321685959,"identity":"6899a047-0250-4a71-ae8e-8e32e55e22ed","order_by":4,"name":"Weijie Zheng","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Weijie","middleName":"","lastName":"Zheng","suffix":""},{"id":321685960,"identity":"89003c1d-a4cb-4d2d-92ac-ee8737892ad2","order_by":5,"name":"Ao Chen","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Ao","middleName":"","lastName":"Chen","suffix":""},{"id":321685961,"identity":"3b4d8bd7-9052-450a-8aed-7c365459be6a","order_by":6,"name":"Senlin Zhu","email":"","orcid":"","institution":"Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Senlin","middleName":"","lastName":"Zhu","suffix":""},{"id":321685962,"identity":"96c80689-ee26-4e52-b2ab-439656dfd5a6","order_by":7,"name":"Tao Shi","email":"","orcid":"","institution":"Northwest A\u0026F University","correspondingAuthor":false,"prefix":"","firstName":"Tao","middleName":"","lastName":"Shi","suffix":""},{"id":321685963,"identity":"8d94abe2-c8a8-4cec-8eaf-694c48dc3ab6","order_by":8,"name":"Yali Hou","email":"","orcid":"","institution":"Chinese Academy of Agricultural Sciences","correspondingAuthor":false,"prefix":"","firstName":"Yali","middleName":"","lastName":"Hou","suffix":""},{"id":321685964,"identity":"1317661b-4055-4aff-a616-07458eb66938","order_by":9,"name":"Fei Wang","email":"","orcid":"","institution":"BGI Research, Shenzhen 518083, China","correspondingAuthor":false,"prefix":"","firstName":"Fei","middleName":"","lastName":"Wang","suffix":""},{"id":321685965,"identity":"412c2b07-70b3-4988-a7f2-ff840a6e675b","order_by":10,"name":"Dong Zou","email":"","orcid":"","institution":"Beijing Institute of Genomics, China National Center for Bioinformation, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Dong","middleName":"","lastName":"Zou","suffix":""},{"id":321685966,"identity":"e8fb6389-327b-4585-b330-559283ea9618","order_by":11,"name":"Yu Song","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Yu","middleName":"","lastName":"Song","suffix":""},{"id":321685967,"identity":"e9b56b0c-168d-4137-bc6b-07398cae0904","order_by":12,"name":"Wen Ye","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Wen","middleName":"","lastName":"Ye","suffix":""},{"id":321685968,"identity":"fd1afe76-a76c-4f07-a955-cd92d914577f","order_by":13,"name":"Aixia Du","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Aixia","middleName":"","lastName":"Du","suffix":""},{"id":321685969,"identity":"1f038c57-5fdc-44bb-aa04-b2a6538ec899","order_by":14,"name":"Yihan Fu","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Yihan","middleName":"","lastName":"Fu","suffix":""},{"id":321685970,"identity":"4b6b2370-e863-4d3d-a34d-ddb8505f428c","order_by":15,"name":"Minghui Jia","email":"","orcid":"","institution":"Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Minghui","middleName":"","lastName":"Jia","suffix":""},{"id":321685971,"identity":"019e81f5-3a8d-4980-adb1-3c206dddd0cd","order_by":16,"name":"Zhonghao Bai","email":"","orcid":"","institution":"Aarhus University","correspondingAuthor":false,"prefix":"","firstName":"Zhonghao","middleName":"","lastName":"Bai","suffix":""},{"id":321685972,"identity":"c6b0d762-2178-486a-b71c-01e8395f466a","order_by":17,"name":"Zhixiang Yuan","email":"","orcid":"","institution":"Chinese Academy of Sciences and China National Center for Bioinformation","correspondingAuthor":false,"prefix":"","firstName":"Zhixiang","middleName":"","lastName":"Yuan","suffix":""},{"id":321685973,"identity":"9403c7ed-91f0-4774-80cb-d655da313adc","order_by":18,"name":"Wan-Sheng Liu","email":"","orcid":"https://orcid.org/0000-0003-1788-7093","institution":"The Pennsylvania State University","correspondingAuthor":false,"prefix":"","firstName":"Wan-Sheng","middleName":"","lastName":"Liu","suffix":""},{"id":321685974,"identity":"f29eaa1c-a7dc-49fc-940e-5bcfba0dcd7c","order_by":19,"name":"Wenbin Tao","email":"","orcid":"","institution":"USDA","correspondingAuthor":false,"prefix":"","firstName":"Wenbin","middleName":"","lastName":"Tao","suffix":""},{"id":321685975,"identity":"e6827fd4-a0f8-409f-828e-de7553e58ad7","order_by":20,"name":"Jayne Hope","email":"","orcid":"","institution":"University of Edinburgh","correspondingAuthor":false,"prefix":"","firstName":"Jayne","middleName":"","lastName":"Hope","suffix":""},{"id":321685976,"identity":"f5541615-ba7d-474e-bbcb-d37548619a6b","order_by":21,"name":"David MacHugh","email":"","orcid":"https://orcid.org/0000-0002-8112-4704","institution":"University College Dublin","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"MacHugh","suffix":""},{"id":321685977,"identity":"3a950005-2a77-4482-aa74-ec1bf58b2273","order_by":22,"name":"John O’Grady","email":"","orcid":"","institution":"University College Dublin","correspondingAuthor":false,"prefix":"","firstName":"John","middleName":"","lastName":"O’Grady","suffix":""},{"id":321685978,"identity":"d0d6cdd4-9e72-4f75-bb82-d1ed86b85f46","order_by":23,"name":"Ole Madsen","email":"","orcid":"","institution":"Wageningen University \u0026 Research","correspondingAuthor":false,"prefix":"","firstName":"Ole","middleName":"","lastName":"Madsen","suffix":""},{"id":321685979,"identity":"633bea24-7463-4af0-9957-cfb63880835d","order_by":24,"name":"Goutam Sahana","email":"","orcid":"https://orcid.org/0000-0001-7608-7577","institution":"Aarhus University","correspondingAuthor":false,"prefix":"","firstName":"Goutam","middleName":"","lastName":"Sahana","suffix":""},{"id":321685980,"identity":"cf3f59f1-3e5d-4d61-863c-67c53474f168","order_by":25,"name":"Yonglun Luo","email":"","orcid":"https://orcid.org/0000-0002-0007-7759","institution":"Aarhus University","correspondingAuthor":false,"prefix":"","firstName":"Yonglun","middleName":"","lastName":"Luo","suffix":""},{"id":321685981,"identity":"31df3baf-99ca-470f-a911-51651f067ee7","order_by":26,"name":"Lin Lin","email":"","orcid":"https://orcid.org/0000-0002-7546-4948","institution":"Aarhus University","correspondingAuthor":false,"prefix":"","firstName":"Lin","middleName":"","lastName":"Lin","suffix":""},{"id":321685982,"identity":"021b0f31-79e4-4603-8076-670809fa2d11","order_by":27,"name":"Cong-jun Li","email":"","orcid":"https://orcid.org/0000-0003-1389-9820","institution":"ARS, USDA","correspondingAuthor":false,"prefix":"","firstName":"Cong-jun","middleName":"","lastName":"Li","suffix":""},{"id":321685983,"identity":"ddcbf4e2-005d-4501-9f03-6b636c0d42eb","order_by":28,"name":"Zexi Cai","email":"","orcid":"https://orcid.org/0000-0002-9579-3415","institution":"Aarhus University","correspondingAuthor":false,"prefix":"","firstName":"Zexi","middleName":"","lastName":"Cai","suffix":""},{"id":321685984,"identity":"1a345a41-0690-4c4c-b15a-cf2c1578e045","order_by":29,"name":"Bingjie Li","email":"","orcid":"https://orcid.org/0000-0002-8528-1889","institution":"Scotland's Rural College (SRUC)","correspondingAuthor":false,"prefix":"","firstName":"Bingjie","middleName":"","lastName":"Li","suffix":""},{"id":321685985,"identity":"e54fe67a-62f8-45b7-b790-cd2b43b736f0","order_by":30,"name":"Lin Liu","email":"","orcid":"","institution":"Beijing Dairy Cattle Center","correspondingAuthor":false,"prefix":"","firstName":"Lin","middleName":"","lastName":"Liu","suffix":""},{"id":321685986,"identity":"8ed4c88a-4283-4be3-899e-69d074d39eb9","order_by":31,"name":"Zhang Zhang","email":"","orcid":"https://orcid.org/0000-0001-6603-5060","institution":"National Genomics Data Center \u0026 CAS Key Laboratory of Genome Sciences and Information, Beijing Institute of Genomics, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Zhang","middleName":"","lastName":"Zhang","suffix":""},{"id":321685987,"identity":"75844aa0-963f-4962-a5cd-c0862fd3c9e2","order_by":32,"name":"Zhu Ma","email":"","orcid":"","institution":"Beijing Dairy Cattle Center","correspondingAuthor":false,"prefix":"","firstName":"Zhu","middleName":"","lastName":"Ma","suffix":""},{"id":321685988,"identity":"8b504887-9b47-4e3c-abb1-0e9e2eeada44","order_by":33,"name":"George E. Liu","email":"","orcid":"https://orcid.org/0000-0003-0192-6705","institution":"Agricultural Research Service","correspondingAuthor":false,"prefix":"","firstName":"George","middleName":"E.","lastName":"Liu","suffix":""},{"id":321685989,"identity":"dd7ee36f-d9f0-4454-9990-43ba0bb3750d","order_by":34,"name":"Yu Jiang","email":"","orcid":"https://orcid.org/0000-0003-4821-3585","institution":"Northwest A\u0026F University","correspondingAuthor":false,"prefix":"","firstName":"Yu","middleName":"","lastName":"Jiang","suffix":""},{"id":321685990,"identity":"4d10e14f-a6bb-4928-831b-d824e6449677","order_by":35,"name":"Hui-Zeng Sun","email":"","orcid":"https://orcid.org/0000-0001-5380-6030","institution":"Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Hui-Zeng","middleName":"","lastName":"Sun","suffix":""},{"id":321685991,"identity":"778a4cb5-ba40-4ae0-bb34-f345351502ea","order_by":36,"name":"Dongxiao Sun","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Dongxiao","middleName":"","lastName":"Sun","suffix":""}],"badges":[],"createdAt":"2024-06-24 17:20:22","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4631710/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4631710/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41588-025-02329-5","type":"published","date":"2025-09-05T04:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":59551532,"identity":"ea8baf0d-f556-4d23-82a1-59826fad5d94","added_by":"auto","created_at":"2024-07-03 06:26:15","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1710732,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCattle cell atlas (CattleCA).\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e, Overview of CattleCA data, analysis, and its application. \u003cstrong\u003eb\u003c/strong\u003e, A schematic diagram of 59 cattle tissues collected in this study. The numbers of samples and cells/nuclei per tissue are shown in parentheses. \u003cstrong\u003ec\u003c/strong\u003e, UMAP visualization of CattleCA colored by different cell types. All the cell types are categorized into seven cell lineages, and cell type annotation and the corresponding cell number are provided in the legend on the right.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/f5022db79a6650fb8a4275a1.png"},{"id":59552023,"identity":"b1675a64-48a5-4cd8-af4a-699d74ee10e8","added_by":"auto","created_at":"2024-07-03 06:34:15","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":900122,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCell heterogeneity within cattle tissue.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea,\u003c/strong\u003e The bar plot at the top left shows the shared cell types across tissues. The ring bar graph shows the numbers of cell types across 59 tissues, with the percentages of unannotated cells marked in red inside the circle. The UMAP plots of three tissues (skin, adrenal gland, and lung) are shown with cell types illustrated by different colors and number of cell types and percentage of unannotated cells are indicated at the bottom right of each of the UMAP plot. \u003cstrong\u003eb,\u003c/strong\u003e The UMAP plot displays subtypes of epithelial cells from three different anatomical components (parenchyma, duct, and cistern) of the mammary gland. \u003cstrong\u003ec,\u003c/strong\u003e The heatmap shows the odds ratio (OR) values of cell subtypes in each anatomical structures of the mammary gland, with OR \u0026gt; 1.5 indicating that the cell subtype is preferentially distributed in the corresponding substructures. \u003cstrong\u003ed,\u003c/strong\u003eHeatmap displays the scaled expression levels of the top 50 cell type-specific marker genes, with selected marker genes highlighted. \u003cstrong\u003ee,\u003c/strong\u003e UMAP visualization of subtypes from germline cells in testis. The names and colors of the cell subtypes correspond to those in Fig. 2b. \u003cstrong\u003ef,\u003c/strong\u003e Pseudotime trajectory analysis of germline cell subtypes with blue shades indicating sorting according to pseudotime values (bottom) and coloring according to cell clusters (top). \u003cstrong\u003eg,\u003c/strong\u003e Expression patterns of TFs over pseudotime. The cell coloring corresponds to Fig. 2e.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/176b84f3fa5931c42b53ca0d.png"},{"id":59552960,"identity":"2d8d8dc7-7388-4c59-a4cb-3e0089416b4d","added_by":"auto","created_at":"2024-07-03 06:42:15","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":2014852,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCellular communication and regulatory networks within tissues and their sex preference.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea,\u003c/strong\u003e Sankey diagram demonstrates the cellular interactions between cell types in 59 tissues. The thickness of the lines represents the strength of the cellular interactions. \u003cstrong\u003eb,\u003c/strong\u003e Cell type-specific transcription factors (TFs). Colored nodes represent cell types, connected to gray nodes representing TFs, with the size of the gray nodes reflecting the number of cell types regulated by each TF. \u003cstrong\u003ec,\u003c/strong\u003e Heatmap shows the average regulatory scores of TFs across 19 common tissues from bulls and cows in this study. The tissue/cell-type specific TFs with sex preference are boxed in red (liver and esophagus), and no sexed bias expression in medulla oblongata. \u003cstrong\u003ed, \u003c/strong\u003eLiver regulon.\u003cstrong\u003e \u003c/strong\u003eThe blue nodes denote TFs, and their connected nodes represent target genes. Pink nodes represent shared target genes, while gray represents non-shared target genes. \u003cstrong\u003ee,\u003c/strong\u003e Significant ligand-receptor pairs originate from hepatocytes and target LSEC/BVEC. Dot color and size show the communication probability and \u003cem\u003ep\u003c/em\u003e-value, respectively. The \u003cem\u003ep\u003c/em\u003e-values are determined through a one-sided permutation test.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/6fe05b7acf99efdf496c0155.png"},{"id":59551540,"identity":"b1142dcf-4f5b-4981-ab9f-6b7f05e72a48","added_by":"auto","created_at":"2024-07-03 06:26:15","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1490802,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eHeterogeneity of immune cells across cattle tissues.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea,\u003c/strong\u003e Bargraph shows the presence of 29 immune cell types across 58 cattle tissues, including both the number (left panel) and proportion (right panel) of immune cells in each tissue are indicated. The scatter plot represents the Shannon entropy values for immune cells in each tissue, wherehigher values indicate a higher abundance of immune cells in that tissue (left) (refer to Supplementary Table 4-1). \u003cstrong\u003eb,\u003c/strong\u003e Pearson correlation analysis of the top 1000 highly variable genes in each cell type of 680,713 immune cells from 48 selected tissues (refer to the row in Supplementary Table 4-1 for tissues with the group column value of ‘selected’, which excludes sub-tissues). \u003cstrong\u003ec,\u003c/strong\u003e t-SNE visualization of annotated cell subtypes of myeloid cells, with cells color-coded according to cell subtypes. \u003cstrong\u003ed,\u003c/strong\u003e Heatmap displays the top five TFs with the highest differences in regulatory specificity scores between each subtype and all other cells in the myeloid cell subtypes. \u003cstrong\u003ee,\u003c/strong\u003ePseudotime trajectory analysis of monocytes and macrophages subtype and cell states colored on the tree. \u003cstrong\u003ef,\u003c/strong\u003e Statistical analysis of monocyte/macrophage subtypes composition in each state. \u003cstrong\u003eg,\u003c/strong\u003ePseudo-heatmap of representative DEGs in differentiation states, and GO enrichment analysis of DEGs re-clustered into four clusters (bottom left). \u003cstrong\u003eh, \u003c/strong\u003eRepresentative pseudotime trajectory of marker genes in different states of microglia.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/881e58e730dfa6194e190462.png"},{"id":59553584,"identity":"4cd29557-1251-4f16-a0f7-4d412a2fc309","added_by":"auto","created_at":"2024-07-03 06:50:15","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1758887,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEpithelial cell heterogeneity and their interaction with immune cells.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea,\u003c/strong\u003e Heatmap of expression for the top 50 DEGs in 24 epithelial cells of 15 selected gastrointestinal tissues, showing significantly enriched KEGG pathways for the clustering module (left), specific TFs (middle), and corresponding tissue anatomical structures and cellular composition (right). \u003cstrong\u003eb,\u003c/strong\u003eScatter plot shows ligand-receptor pairs between goblet cells (GCs) and seven types of immune cells, which are shared across six intestinal segments. \u003cstrong\u003ec, \u003c/strong\u003eThe schematic diagram shows the cellular communication model shared between GCs and immune cells across six intestinal segments. \u003cstrong\u003ed,\u003c/strong\u003e Scatter plot shows the gene expression of ligand-receptor pairs shared across multiple tissues in different cell types. \u003cstrong\u003ee,\u003c/strong\u003e t-SNE visualization of seven GC subtypes. Cells are color-coded according to cell subtypes. \u003cstrong\u003ef, \u003c/strong\u003eScatter plot shows GO terms enriched by DGEs among seven GC subtypes. \u003cstrong\u003eg,\u003c/strong\u003e Cellular communication analysis between seven GC subtypes and immune cells, where the width of intercellular connections represents the strength of communication. \u003cstrong\u003eh-i, \u003c/strong\u003eHierarchical diagram (left) displays the cellular communication strength between cell types, with solid circles and hollow circles representing the source and target, respectively. The size of the circle is proportional to the number of cells in each cell group. The edge color is consistent with the source. The thicker the line, the stronger the communication signal. Heatmap (right) shows the signal transduction of cell types, identifying the main senders, receivers, mediators, and influencers in the intercellular communication network by calculating the network centrality index of each cell type.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/61824900b43a8c8fe141f14b.png"},{"id":59551544,"identity":"8aa9b56d-a24e-4c84-b2c7-9273b66deb35","added_by":"auto","created_at":"2024-07-03 06:26:15","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1155808,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMonogenic disorder related to cell types.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea, \u003c/strong\u003eCell-type correlation of gene expression within cell type-specific genes (tau \u0026gt; 0.85). \u003cstrong\u003eb, \u003c/strong\u003eCorrelations between seven cell lineages and ten disorder domains. The dot sizes represent Log\u003csub\u003e2\u003c/sub\u003eOR values, and the color represents significant levels. Gene numbers for each cell lineage and disorder domain are denoted in parentheses. \u003cstrong\u003ec, \u003c/strong\u003eExpression level of nine epithelial-specific skin disorder genes in different cell lineages. The star symbol on the top of the violin represents a significant expression difference between this cell lineage and epithelial cells. \u003cstrong\u003ed,\u003c/strong\u003e Expression levels of nine epithelial-specific skin disorder genes in different epithelial cell types. The skin disorders corresponding to the genes are denoted in parentheses. The dot size represents log\u003csub\u003e10\u003c/sub\u003eCPM values and the color represents the z-score of genes in each cell type. \u003cstrong\u003ee,\u003c/strong\u003e Putative cellular communication between epithelial cells and other cell types implicating skin disorder genes. Cell types (inner color) in different tissues (outer color) are connected by putative interactions (dotted lines) between a ligand (left square) expressed in one epithelial cell and a receptor (right square) expressed in the other cell types. Bold italic indicates a skin disorder-related gene. \u003cstrong\u003ef, \u003c/strong\u003eThe expression distribution of \u003cem\u003eLAMA3\u003c/em\u003e in the placenta primarily occurs in uninucleate trophoblast cells, which is associated with the skin disorder epidermolysis bullosa (EB). \u003cstrong\u003eg,\u003c/strong\u003e Re-clustering and trajectory analysis of uninucleate trophoblast cells (UTCs) in the placenta. Shown is the distribution of \u003cem\u003eLAMA3\u003c/em\u003e expression (log\u003csub\u003e10\u003c/sub\u003eCPM) in different cell subtypes through the trajectory line. \u003cstrong\u003eh, \u003c/strong\u003ePutative cellular communication between UTC subtypes and other cell types in the placenta. Cell types (inner color) in the placenta (outer color) are connected by putative interactions (dotted lines) between a ligand (left square) expressed in one UTC subtype and a receptor (right square) expressed in the other cell types. \u003cstrong\u003ei,\u003c/strong\u003e Expression levels of muscle disorder genes in different muscle cell types. The dot size represents log\u003csub\u003e10\u003c/sub\u003eCPM values and the color represents the z-score of genes in each cell type. The muscle disorders corresponding to the genes are denoted in parentheses. \u003cstrong\u003ej, \u003c/strong\u003eExpression distribution of \u003cem\u003eMYBPC1\u003c/em\u003e in the esophagus. The muscle disorder corresponding to the gene is denoted in parentheses. \u003cstrong\u003ek,\u003c/strong\u003e Re-clustering and trajectory analysis of skeletal muscle cells (SMCs) in the esophagus. Shown is the distribution of \u003cem\u003eMYBPC1\u003c/em\u003e expression (log\u003csub\u003e10\u003c/sub\u003eCPM) in different cell subtypes through the trajectory line.\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/a3c4b145ae5e646a3e52d6bb.png"},{"id":59551537,"identity":"0b92236d-7461-4b5f-9631-7b076cc2cb6a","added_by":"auto","created_at":"2024-07-03 06:26:15","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":1029929,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEnrichments between cell types and complex traits.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e, Correlations between cell lineages and complex traits. Heatmap represents the significant level of each cell lineage with traits. Gray squares on the top represent trait groups. \u003cstrong\u003eb-c\u003c/strong\u003e, Correlations between milk production (b) and sperm (c) traits and cell types. Each circle represents a cell-type-trait association. The x-axis represents cell lineages sorted alphabetically. \u003cstrong\u003ed-f\u003c/strong\u003e, Correlation between cell types and traits in tissues. Violin plot represents the trait-relevant score (TRS) for each cell type with the corresponding trait. The red line represents the significant level (-log\u003csub\u003e10\u003c/sub\u003ep-value) for each cell type with the corresponding trait. \u003cstrong\u003eg\u003c/strong\u003e, Active KEGG pathways in the top significant trait relevant cell types. The dot size represents the significant level (-log\u003csub\u003e10\u003c/sub\u003e\u003cem\u003ep\u003c/em\u003e-value) of pathway activity, and the dot color represents the significant level of enrichment between active pathway genes and top trait relevant genes in the given cell type.\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/37b6843ef3124fdedbf2caf7.png"},{"id":59551546,"identity":"7809ad17-0495-4f23-8c3d-b0769f56bd60","added_by":"auto","created_at":"2024-07-03 06:26:16","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":812079,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCross-species comparisons and disease associations.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e, MetaNeighbor AUROC values calculated using shared genes of the same cell types in cattle and humans. \u003cstrong\u003eb-c\u003c/strong\u003e, Regulon specificity score (RSS) of TFs in immune cells (\u003cstrong\u003eb\u003c/strong\u003e) and epithelial cells (\u003cstrong\u003ec\u003c/strong\u003e). Scatter plot represents TFs that were identified in both cattle and humans, in which, the x-axis shows the RSS of each TF in cattle and the y-axis shows the RSS in humans. Representative TFs were considered to be significant (RSS \u0026gt; 0.25) and labeled in color. The r represents the correlation coefficient of RSS on the two species, and the \u003cem\u003ep\u003c/em\u003e-value represents the significance of the correlation coefficient. \u003cstrong\u003ed-g, \u003c/strong\u003eCellular communication in the small intestine (\u003cstrong\u003ed-e)\u003c/strong\u003e and liver (\u003cstrong\u003ef-g\u003c/strong\u003e) between cattle and humans. The network diagram shows the analysis of cellular communication between cattle and humans, and the column stacking diagram represents the proportion of the two species in the signaling pathway. Red represents pathways significantly upregulated in cattle, blue represents pathways significantly upregulated in humans, and black represents pathways that are not significant in either cattle or humans. \u003cstrong\u003eh-i,\u003c/strong\u003e Heritability enrichment of cell types for orthologous marker genes with MS (h) and IBD (i). The y-axis represents the significant level (-log\u003csub\u003e10\u003c/sub\u003e\u003cem\u003ep\u003c/em\u003e-value) of heritability enrichment and the x-axis represents all cell types identified in the corresponding cattle tissue.\u003c/p\u003e","description":"","filename":"image8.png","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/d24597b4a28b6f3229b903e3.png"},{"id":90711784,"identity":"aed09f12-8580-4e87-a2a4-071089d0ac2c","added_by":"auto","created_at":"2025-09-06 07:09:27","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":14131151,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/7eca3072-d5f4-49fc-be31-1ea26829942c.pdf"},{"id":59551553,"identity":"651b18b0-e004-41bf-8852-5ebc39c0fb59","added_by":"auto","created_at":"2024-07-03 06:26:16","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":36226133,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"CattleCellAtalsSupplementaryinformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/42034c1eef5e82983b958f27.docx"},{"id":59552021,"identity":"d24451f4-8c9f-4983-95bb-54c90e100fc7","added_by":"auto","created_at":"2024-07-03 06:34:15","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":15609,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable11.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/c46d20354f46e7a240ebf5d1.xlsx"},{"id":59551534,"identity":"665a1d24-8ec5-4697-a6f2-3e7621f89216","added_by":"auto","created_at":"2024-07-03 06:26:15","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":10439,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable12.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/a202e8c1961376ce6b3766bd.xlsx"},{"id":59551535,"identity":"49fd1b16-9ccb-4671-9ecd-7baf1f3574df","added_by":"auto","created_at":"2024-07-03 06:26:15","extension":"xlsx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":17560,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable13.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/6a17cd159bf942a92d1b03c6.xlsx"},{"id":59551559,"identity":"79ee9274-13fc-43f7-9cee-ac72169209af","added_by":"auto","created_at":"2024-07-03 06:26:18","extension":"xlsx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":125124350,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable14.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/c4b3a0173c44aca0f5e31db2.xlsx"},{"id":59551554,"identity":"7d372b6c-d9e4-463e-81c4-290e230d3e89","added_by":"auto","created_at":"2024-07-03 06:26:16","extension":"xlsx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":254838,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable21.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/c276c404fbca91c012a5ef1e.xlsx"},{"id":59552028,"identity":"78662e17-923c-4835-a492-8b78b1478db1","added_by":"auto","created_at":"2024-07-03 06:34:16","extension":"xlsx","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":4875545,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable31.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/1ce4212fa08693e50b6da95b.xlsx"},{"id":59552027,"identity":"916ab230-d95a-4bd5-b07a-bd87bfefdd10","added_by":"auto","created_at":"2024-07-03 06:34:16","extension":"xlsx","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":20783,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable41.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/3853ade8e0456c7809371102.xlsx"},{"id":59551549,"identity":"33b78f70-588c-4f81-a921-3cf70f2ef610","added_by":"auto","created_at":"2024-07-03 06:26:16","extension":"xls","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":21504,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable42.xls","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/7cb3462c54ee4172e479e203.xls"},{"id":59552026,"identity":"9189bcfb-f28d-4cf4-9bf3-d924bcfd1a10","added_by":"auto","created_at":"2024-07-03 06:34:15","extension":"xlsx","order_by":10,"title":"","display":"","copyAsset":false,"role":"supplement","size":12499,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable51.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/e5dc9dfc26089febc8863105.xlsx"},{"id":59551548,"identity":"02148ee3-1721-4461-822a-d9ca9c5754d1","added_by":"auto","created_at":"2024-07-03 06:26:16","extension":"xlsx","order_by":11,"title":"","display":"","copyAsset":false,"role":"supplement","size":76320,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable52.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/66a7deaf4f4c10ae1637abd1.xlsx"},{"id":59551543,"identity":"69b17844-4879-4994-9ff8-0d2e8f2e8b7a","added_by":"auto","created_at":"2024-07-03 06:26:15","extension":"xlsx","order_by":12,"title":"","display":"","copyAsset":false,"role":"supplement","size":21019,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable61.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/570009a52b9850563fdbb483.xlsx"},{"id":59551557,"identity":"c607ddb2-5e07-4781-a72d-d7e6e7571f0d","added_by":"auto","created_at":"2024-07-03 06:26:17","extension":"xlsx","order_by":13,"title":"","display":"","copyAsset":false,"role":"supplement","size":47479927,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable62.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/085ceef0d7cca2e8812898e2.xlsx"},{"id":59552029,"identity":"921cb822-374d-45b8-a02f-4256c8b27359","added_by":"auto","created_at":"2024-07-03 06:34:17","extension":"xlsx","order_by":14,"title":"","display":"","copyAsset":false,"role":"supplement","size":111845,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable63.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/979e732b81fc0c36e70ce2dc.xlsx"},{"id":59552962,"identity":"ec83a9a6-e345-43c4-902e-79b88ec3305b","added_by":"auto","created_at":"2024-07-03 06:42:15","extension":"xlsx","order_by":15,"title":"","display":"","copyAsset":false,"role":"supplement","size":33896,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable71.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/9c40c83cf4acd64ad928fd55.xlsx"},{"id":59551545,"identity":"eac38a42-6b09-41e3-86af-c3d7c892c31c","added_by":"auto","created_at":"2024-07-03 06:26:15","extension":"xlsx","order_by":16,"title":"","display":"","copyAsset":false,"role":"supplement","size":4465777,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable72.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/55000dc8eab5a75cfe0c0c4c.xlsx"},{"id":59551551,"identity":"80ea31e1-cd11-4256-a3c1-060bf354e982","added_by":"auto","created_at":"2024-07-03 06:26:16","extension":"xlsx","order_by":17,"title":"","display":"","copyAsset":false,"role":"supplement","size":17736,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable81.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/10d1b60ce37055be1e7216e9.xlsx"},{"id":59551547,"identity":"2f4b55ac-e13a-4d6e-be4d-8361d73d3130","added_by":"auto","created_at":"2024-07-03 06:26:16","extension":"xlsx","order_by":18,"title":"","display":"","copyAsset":false,"role":"supplement","size":26165,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable82.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/a92ab7f17af00e8a24c010a5.xlsx"},{"id":59551556,"identity":"bddaf9be-0b49-444c-8198-5a39b570580d","added_by":"auto","created_at":"2024-07-03 06:26:17","extension":"xlsx","order_by":19,"title":"","display":"","copyAsset":false,"role":"supplement","size":71312,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable83.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4631710/v1/4d1b45bae7442701ec4b20cf.xlsx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Cattle Cell Atlas: a multi-tissue single cell expression repository for advanced bovine genomics and comparative biology","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCattle, domesticated over 10,000 years ago\u003csup\u003e1\u0026ndash;3\u003c/sup\u003e, play a crucial role in converting indigestible fiber feed into protein-rich food to humans such as beef and milk, which are essential for humans nutrition and health. To meet increasing global demand for safe animal food products while minimizing the production-associated negative impacts on animal welfare and the environment (e.g., greenhouse gas emissions and zoonotic diseases)\u003csup\u003e4\u0026ndash;7\u003c/sup\u003e, it is essential to understand the genetic and molecular mechanisms underlying various phenotypes of economic and ecological importance in cattle.\u003c/p\u003e \u003cp\u003eAs of 28 April, 2024, 191,181 genomic loci have been reported to be associated with over 600 different complex traits in cattle\u003csup\u003e8\u003c/sup\u003e. A large proportion of these variants reside in non-coding genomic regions, have small individual effects on phenotypic variation, and influence complex traits \u003cem\u003evia\u003c/em\u003e modulation of gene regulation. In this regard, numerous studies have explored the molecular mechanisms underlying complex traits at the tissue expression levelthrough integrating genome-wide association study (GWAS) data sets and molecular quantitative trait loci (molQTL)\u003csup\u003e9\u0026ndash;12\u003c/sup\u003e. For instance, the Cattle Genotype-Tissue Expression (CattleGTEx) project\u003csup\u003e13\u003c/sup\u003e linked gene expression in over 20 tissues with 43 significant economically important traits, provides valuable insights into their gene regulatory mechanisms. However, tissues are generally heterogenous mixtures of distinct cell types and states. As a consequence of the rapid development of single-cell sequencing technology, single-cell transcriptome atlases have been constructed for many organisms, including humans\u003csup\u003e14,15\u003c/sup\u003e, mice\u003csup\u003e16\u003c/sup\u003e, fruit flies\u003csup\u003e17\u003c/sup\u003e, zebrafish\u003csup\u003e18\u003c/sup\u003e, axolotl\u003csup\u003e19\u003c/sup\u003e, and pigs\u003csup\u003e20\u003c/sup\u003e. However, in cattle, previous studies on single-cell transcriptomic studies were limited to specific tissue types including the rumen\u003csup\u003e21\u0026ndash;23\u003c/sup\u003e, peripheral blood \u003csup\u003e24,25\u003c/sup\u003e, skeletal muscle\u003csup\u003e26\u003c/sup\u003e, and the digestive system\u003csup\u003e27,28\u003c/sup\u003e. Therefore, there is an urgent need to comprehensively catalogue different bovine cell types and states across various tissue types and biological contexts, which will substantially contribute to our understanding of the genetic and molecular architecture underlying numerous phenotypes in cattle.\u003c/p\u003e \u003cp\u003eIn this study, we build a comprehensive Cattle Cell Atlas (CattleCA) for the livestock research community by generating and analyzing single-cell/nucleus RNA sequencing (sc/snRNA-seq) data from 1,793,854 cells across 59 tissue types in 15 animals, spanning both sexes and three developmental stages (fetus, calves, and adults; \u003cb\u003eFig.\u0026nbsp;1)\u003c/b\u003e. We characterize 131 distinct cell types and assess the cellular heterogeneity in terms of gene expression, transcription factor regulation, and intra- and inter-tissue cellular communications. Leveraging the CattleCA, we highlight specific cell types/states associated with monogenic conditions and complex traits. Furthermore, we explored the evolutionary conservation of the transcriptome between cattle and humans at single-cell level, revealing shared cellular mechanisms underlying complex traits and diseases in humans. The CattleCA (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://ngdc.cncb.ac.cn/cattleca/\u003c/span\u003e\u003cspan address=\"https://ngdc.cncb.ac.cn/cattleca/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) is thus of fundamental significance to serve as an invaluable resource for cattle genetics/genomics, immunology, precision breeding, and comparative biology.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eThe scope of the Cattle Cell Atlas\u003c/h2\u003e \u003cp\u003eWe analyzed the transcriptome data from a total of 1,506,438 single cells and 287,416 single nuclei (hereinafter referred to as cells) from one fetus, four calves, and ten adults (\u003cb\u003eSupplementary Fig.\u0026nbsp;1\u0026ndash;1 and Supplementary Table\u0026nbsp;1\u0026ndash;1\u003c/b\u003e). These cells included 234,802 from males and 1,559,052 from females. By integrating all these cells using Harmony\u003csup\u003e29\u003c/sup\u003e, we annotated 131 major cell types based on canonical marker genes, representing seven distinct cell lineages: immune (n\u0026thinsp;=\u0026thinsp;679,021), endothelial (n\u0026thinsp;=\u0026thinsp;308,878), epithelial (n\u0026thinsp;=\u0026thinsp;268,626), stromal (n\u0026thinsp;=\u0026thinsp;240,771), nerve (n\u0026thinsp;=\u0026thinsp;182,450), muscle (n\u0026thinsp;=\u0026thinsp;81,438), and germline (n\u0026thinsp;=\u0026thinsp;4,404) cells (\u003cb\u003eFig.\u0026nbsp;1c; Supplementary Fig.\u0026nbsp;1-2a and Supplementary Table\u0026nbsp;1\u0026ndash;2, 3\u003c/b\u003e). These cells clustered well based on cell lineage types rather than processing methods, sequencing platforms and tissue types (\u003cb\u003eFig.\u0026nbsp;1c and Supplementary Fig.\u0026nbsp;1-2b-d\u003c/b\u003e), with cell type abundance ranging from 33 for circulating epithelial cells to 233,497 for blood vascular endothelial cells (BVECs) (\u003cb\u003eFig.\u0026nbsp;1c\u003c/b\u003e). On average, 12 distinct cell types were identified per tissue, ranging from 7 in the oviduct to 21 in the ileum (\u003cb\u003eFig.\u0026nbsp;2a; Supplementary Table\u0026nbsp;2\u0026thinsp;\u0026minus;\u0026thinsp;1; and Supplementary Fig.\u0026nbsp;2\u0026thinsp;\u0026minus;\u0026thinsp;1 and 2\u0026ndash;2)\u003c/b\u003e. Among the 131 annotated cell types, 67 were observed only in one tissue type, whereas immune, endothelial and epithelial cells were found across 58, 49, and 41 tissues, respectively (\u003cb\u003eFig.\u0026nbsp;2a; Supplementary Fig.\u0026nbsp;1-2e and Supplementary Table\u0026nbsp;2\u0026thinsp;\u0026minus;\u0026thinsp;1\u003c/b\u003e). For instance, BVECs were ubiquitously present in 45 tissues, whereas α and β cells were exclusively detected in the pancreas. Furthermore, cell cycle analysis indicated that cells from germline lineage were predominantly enriched in the G2/M phase, reflecting their active growth and preparation for DNA replication. Conversely, epithelial, muscle, stromal, and endothelial cells were primarily enriched in the G0/G1 phase, indicating a quiescent state with no active division (\u003cb\u003eSupplementary Fig.\u0026nbsp;1-2f and Supplementary Table\u0026nbsp;1\u0026ndash;4\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eCellular heterogeneity in mammary gland and testis\u003c/h2\u003e \u003cp\u003eTo study the cellular heterogeneity within tissues, we took mammary gland and testis as examples due to their potential importance in milk production and male fertility traits of economic value in dairy cattle. We identified 10 epithelial cell subtypes (ME0-ME9) from 4,819 cells isolated from mammary parenchyma, duct, and cistern (\u003cb\u003eFig.\u0026nbsp;2b\u003c/b\u003e). These cell subtypes had diverse distribution patterns and functional characteristics across anatomical components of the mammary gland (\u003cb\u003eFig.\u0026nbsp;2c, d\u003c/b\u003e). The luminal secretory (LumSec) cell subtypes (ME0, ME1, ME2, ME3, ME4, and ME5) were associated with milk biosynthesis, expressing genes like \u003cem\u003eCSN2\u003c/em\u003e and \u003cem\u003eLALBA\u003c/em\u003e, and were enriched in the mammary cistern (M1, ME2, ME3, and ME4) and mammary parenchyma (ME0 and ME5; \u003cb\u003eFig.\u0026nbsp;2c, d and Supplementary Fig.\u0026nbsp;2-3a\u003c/b\u003e). Genes with high expression in M1, ME2, ME3, and ME4 were significantly enriched in lipid metabolism, epithelial cell differentiation, cell morphogenesis, and immune defense, respectively (\u003cb\u003eSupplementary Fig.\u0026nbsp;2-3b\u003c/b\u003e). The ME0 and ME5 subtypes, displaying upregulated genes such as \u003cem\u003eCXCL8\u003c/em\u003e and \u003cem\u003eCCL3\u003c/em\u003e (\u003cb\u003eSupplementary Fig.\u0026nbsp;2-3a\u003c/b\u003e), were associated with host defense by the epithelial cells in chemotaxis and immune cell recruitment\u003csup\u003e30\u003c/sup\u003e. Conversely, luminal hormone-responsive (LumHR) cell subtypes, ME8 and ME9, were primarily derived from mammary ducts (\u003cb\u003eFig.\u0026nbsp;2c\u003c/b\u003e) and involved in ductal development\u003csup\u003e31\u003c/sup\u003e through paracrine EGFR-ligands, \u003cem\u003eAREG\u003c/em\u003e, \u003cem\u003eTGFA\u003c/em\u003e, \u003cem\u003eBTC\u003c/em\u003e, and \u003cem\u003eHBEGF\u003c/em\u003e (\u003cb\u003eSupplementary Fig.\u0026nbsp;2-3c\u003c/b\u003e), cross-acting on myoepithelial cells (ME7).\u003c/p\u003e \u003cp\u003eIn testis, we categorized 3,958 germline cells into 12 cell subtypes, representing the progression from spermatogonia (S5 and S7), through spermatocytes (S0, S3, S6, S9, and S10), to spermatids (S1, S2, S4, S8, and S11; \u003cb\u003eFig.\u0026nbsp;2e and Supplementary Fig.\u0026nbsp;2\u0026ndash;4\u003c/b\u003e). Pseudotime analysis validated the developmental trajectory of gametogenesis (\u003cb\u003eFig.\u0026nbsp;2f\u003c/b\u003e). Interestingly, spermatocytes displayed a bifurcated differentiation pattern from the early stage (S3) to the late stage (S0) (\u003cb\u003eFig.\u0026nbsp;2f\u003c/b\u003e), consistent with previous findings in yaks or cattle-yak hybrids\u003csup\u003e32\u003c/sup\u003e, but not in humans\u003csup\u003e33\u003c/sup\u003e. Furthermore, we identified nine pivotal transcription factors (TFs) orchestrating the gametogenesis process (\u003cb\u003eFig.\u0026nbsp;2g).\u003c/b\u003e For instance, BCLAF1 and HLTF exhibited reduced expression levels during the spermatogonial transition (S5 to S7), suggesting putative roles for these TFs in sustaining spermatogonial stemness. YY1, known to influence chromosome double-strand breaks\u003csup\u003e34\u003c/sup\u003e, exhibited a peak expression in S3. CREM and MBD1 demonstrated increased expression from late spermatocytes onwards, which might influence sperm development\u003csup\u003e35,36\u003c/sup\u003e. As crucial TFs for spermiogenesis, the expression patterns of ELF2, RFX2, RFX4, and CUX1 remained constant during the process from late spermatogonia to the maturation of spermatids (\u003cb\u003eSupplementary Fig.\u0026nbsp;2\u0026ndash;5\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eIntercellular communication and transcriptional regulation within tissues\u003c/h2\u003e \u003cp\u003eCellular interactions play essential roles in shaping the functions of multicellular organisms\u003csup\u003e37\u003c/sup\u003e. Using CellChat\u003csup\u003e38\u003c/sup\u003e, we found that most cellular communications occurred within cell lineages, with immune and epithelial cell lineages showing higher communication (\u003cb\u003eSupplementary Fig.\u0026nbsp;3\u0026thinsp;\u0026minus;\u0026thinsp;1\u003c/b\u003e). Stromal cells emitted the most signals, essential for maintaining tissue homeostasis (\u003cb\u003eSupplementary Fig.\u0026nbsp;3\u0026thinsp;\u0026minus;\u0026thinsp;1\u003c/b\u003e). The intensity and pattern of intercellular interactions were tissue-specific, with the uterine horn showing the strongest and the retina the weakest interactions (\u003cb\u003eFig.\u0026nbsp;3a\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eTo dissect gene regulatory networks at the cell type level, we conducted a systematic analysis using pySCENIC\u003csup\u003e39\u003c/sup\u003e, identifying 14,094 regulons for 980 TFs across 59 tissues (\u003cb\u003eSupplementary Table\u0026nbsp;3\u0026thinsp;\u0026minus;\u0026thinsp;1\u003c/b\u003e). Among these TFs, 710 showed cell-type-specific activity across 130 cell types (excluding proliferative cells), potentially playing pivotal roles in determining cellular identity and fate. For example, GATA1, SPI, and HNF4A are crucial for mast cell, macrophage, and hepatocyte differentiation, respectively (\u003cb\u003eFig.\u0026nbsp;3b\u003c/b\u003e). TFs regulating a broader spectrum of cell types exhibited stronger DNA sequence constraints (\u003cb\u003eSupplementary Fig.\u0026nbsp;3-2a, b\u003c/b\u003e). For instance, HOXA2, which influences regulatory processes across 35 distinct cell types, had the highest PhastCons and phyloP scores, within its target genes which were significantly enriched in the cytoplasm, negative regulation of expression, and identical protein binding (\u003cb\u003eSupplementary Fig.\u0026nbsp;3-2a-c\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eTo investigate the impact of sex on transcriptional regulation, we compared male and female regulon activity scores (RAS) across 19 tissues with data for both sexes (\u003cb\u003eFig.\u0026nbsp;3c; Supplementary Fig.\u0026nbsp;1\u0026ndash;1 and Supplementary Fig.\u0026nbsp;3\u0026ndash;3\u003c/b\u003e). We identified one tissue exhibiting no sex-biased regulation (medulla oblongata), four tissues displaying sex-biased regulation (i.e., heart, kidney, esophagus, and skin), and the remaining 14 tissues manifesting partial sex-biased regulation limited to specific cell types (\u003cb\u003eFig.\u0026nbsp;3c and Supplementary Fig.\u0026nbsp;3\u0026ndash;3\u003c/b\u003e). Taking the liver as an example, three TFs, HNF4A, NR1I3, and NFIA, exhibited high regulatory activity specifically in female hepatocytes, with their target genes (e.g., \u003cem\u003eAGMO, KHK\u003c/em\u003e, and \u003cem\u003eFTCD\u003c/em\u003e; \u003cb\u003eFig.\u0026nbsp;3d\u003c/b\u003e) participating in fatty acid metabolism\u003csup\u003e40\u0026ndash;42\u003c/sup\u003e. Similarly, SOX18 specifically regulates female liver sinusoidal endothelial cells (LSECs) and BVECs, targeting signaling pathway-related genes\u003csup\u003e43,44\u003c/sup\u003e, such as \u003cem\u003eHPCAL1\u003c/em\u003e and \u003cem\u003eRAMP3\u003c/em\u003e (\u003cb\u003eFig.\u0026nbsp;3d\u003c/b\u003e). The four TFs and their five target genes detailed above also have higher expression in the female human liver compared to the male liver (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://gtexportal.org/home/\u003c/span\u003e\u003cspan address=\"https://gtexportal.org/home/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). We further analyzed the cellular communication between hepatocytes and the two endothelial cells in females and found that their ligand-receptor pairs primarily participate in the VEGF signaling pathway, including VEGFA-VEGFR1, VEGFA-VEGFR1R2, and VEGFA-VEGFR2 (\u003cb\u003eFig.\u0026nbsp;3e\u003c/b\u003e). The VEGF pathway plays essential roles in influx and efflux of substances such as lipoproteins and chylomicron remnants\u003csup\u003e45,46\u003c/sup\u003e. Taken together, our findings indicate that these sex-biased regulatory factors in the liver may play key roles in lipid metabolism and milk production in lactating dairy cows.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eAntigen-presenting immune cell heterogeneity\u003c/h2\u003e \u003cp\u003eWe identified 777,873 immune cells from 58 tissues (excluding the retina), categorizing them into 29 cell types (\u003cb\u003eFig.\u0026nbsp;4a and Supplementary Table\u0026nbsp;4\u0026thinsp;\u0026minus;\u0026thinsp;1\u003c/b\u003e). The mammary gland (96,865), liver (55,369), and PBMC (46,078) exhibited the highest number of immune cells. While macrophages were found across 38 tissues, double-negative T cells, Kupffer cells, and plasmacytoid dendritic cells (pDCs) were detected only in thymus, liver, and PBMC, respectively (\u003cb\u003eFig.\u0026nbsp;4a\u003c/b\u003e). According to gene expression similarities, lymphocytes and myeloid cells (MCs) were well-clustered, indicating their similar functional and lineage-specific features, whereas microglia, neutrophils, and pDCs displayed a weak correlation with other immune cells (\u003cb\u003eFig.\u0026nbsp;4b\u003c/b\u003e). Plasma cells and erythroid cells showed pronounced heterogeneity compared to other immune cells (\u003cb\u003eFig.\u0026nbsp;4b\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eAntigen-presenting cells (APCs) are central to the induction of adaptive immunity, mediating immune responses by processing and presenting antigens to lymphocytes\u003csup\u003e47,48\u003c/sup\u003e. Within the myeloid cell cluster, a key component of APCs, we identified 169,610 cells from 44 tissues, which were further categorized into six macrophage subtypes (MA0-MA5), three monocyte subtypes (MO0-MO2), and four dendritic cell subtypes (DC0-DC3) (\u003cb\u003eFig.\u0026nbsp;4c and Supplementary Fig.\u0026nbsp;4-1a\u003c/b\u003e). A total of 135,186 macrophages were present in 40 tissues, making macrophages an appropriate model for exploring cellular heterogeneity across tissue microenvironments. We found the heterogeneity for the following aspects: 1) unique transcription factor regulation was observed across cell subtypes (\u003cb\u003eFig.\u0026nbsp;4d\u003c/b\u003e); elevated expression levels of \u003cem\u003eIRF7\u003c/em\u003e and \u003cem\u003eIRF8\u003c/em\u003e in MA1 potentially sustain macrophage functionality and development, while \u003cem\u003eBHLHE41\u003c/em\u003e, highly expressed in MA0, may inhibit the replacement of exogenous macrophages and enhance tissue-specific macrophage characteristics; 2) varied antigen-presenting ability was evident among cell subtypes, exemplified by MA1, primarily distributed in the intestine, exhibiting the strongest MHC-II antigen-presenting score (APS), contrasting with MA4 in the liver, which displayed the weakest APS (\u003cb\u003eSupplementary Fig.\u0026nbsp;4-1b, c and Supplementary Table\u0026nbsp;4\u0026thinsp;\u0026minus;\u0026thinsp;2\u003c/b\u003e); 3) a Meta-neighbor analysis revealed six functional modules (\u003cb\u003eSupplementary Fig.\u0026nbsp;4-2a, c-j\u003c/b\u003e), where Module 6, characterized by upregulated genes associated with intracellular signal transduction and primarily composed of MA0 from the brain, accounted for 92.4% of the cells in that tissue (\u003cb\u003eSupplementary Fig.\u0026nbsp;4-2a, b\u003c/b\u003e); 4) across 20 tissues, macrophages exhibited a propensity towards the classically activated M1 phenotype, predominantly in the reproductive and nervous systems; conversely, 19 tissues exhibited a tendency towards alternatively activated M2 macrophages which was observed in tissues such as the intestine (\u003cb\u003eSupplementary Fig.\u0026nbsp;4-1e\u003c/b\u003e); 5) pseudotime trajectory analysis revealed a bifurcated differentiation pattern of macrophages derived from monocytes (\u003cb\u003eFig.\u0026nbsp;4e\u003c/b\u003e); on the right branch, MA1 and MA2, derived from more developed monocytes (MO2), exhibited activated genes related to transmembrane transport. Conversely, the left branch depicted gradual differentiation of MA4, MA3, and MA0, with activated genes enriched in neural development, including microglia markers such as \u003cem\u003eP2RY12\u003c/em\u003e, \u003cem\u003eGPC5\u003c/em\u003e, \u003cem\u003eCALCR\u003c/em\u003e, and \u003cem\u003eCX3CR1\u003c/em\u003e (\u003cb\u003eFig.\u0026nbsp;4e-h and Supplementary Fig.\u0026nbsp;4-1d\u003c/b\u003e). Since macrophages can be derived from either circulating monocytes or embryonic progenitor cells\u003csup\u003e49\u0026ndash;51\u003c/sup\u003e, it is suggested that MA1 and MA2 might be derived from monocytes, while MA0 (brain accounting for 96.24%), MA3 (64.46% liver, 30.58% spleen) and MA4 (96.24% liver) with tissue specificity might be derived from embryonic progenitor cells.\u003c/p\u003e \u003cp\u003eSimilarly, we investigated the heterogeneity of B cells across different tissues, as they can also act as APCs\u003csup\u003e52,53\u003c/sup\u003e. We categorized 76,410 B cells from 29 tissues into 8 subtypes (\u003cb\u003eSupplementary Fig.\u0026nbsp;4-3a-c)\u003c/b\u003e. Transcripts for key TFs associated with diverse B cell differentiation fates, such as YY1, BCL6, and BACH2, were detected (\u003cb\u003eSupplementary Fig.\u0026nbsp;4-3d\u003c/b\u003e), which play crucial roles in developing germinal center B cells\u003csup\u003e54,55\u003c/sup\u003e. \u003cem\u003eATF4\u003c/em\u003e and \u003cem\u003eXBP1\u003c/em\u003e, highly expressed in both P0 and P2, were found to regulate B cell differentiation into plasma cells\u003csup\u003e56,57\u003c/sup\u003e. Plasma cells exhibited a significantly lower APS for MHC-II compared to B cells, which may be attributed to their origin from antigen-stimulated B cells and their primary antibody production function and secretion rather than antigen presentation\u003csup\u003e58\u003c/sup\u003e (\u003cb\u003eSupplementary Fig.\u0026nbsp;4-3e\u003c/b\u003e). Pseudotime trajectory analysis demonstrated the differentiation process from B cells to plasma cells (\u003cb\u003eSupplementary Fig.\u0026nbsp;4-3f, g\u003c/b\u003e). During this process, markers specific to B0/B1 subtypes, such as \u003cem\u003eEBF1, BACH2\u003c/em\u003e, and \u003cem\u003eCCR7\u003c/em\u003e, were upregulated in the early stages of differentiation, followed by the activation of intermediate B cell subtype markers, including \u003cem\u003eCD79A\u003c/em\u003e, \u003cem\u003eCD79B\u003c/em\u003e, \u003cem\u003eBCL6\u003c/em\u003e, and \u003cem\u003eNEIL1\u003c/em\u003e. Finally, the expression of plasma cell markers like \u003cem\u003eJCHAIN\u003c/em\u003e and \u003cem\u003eMZB1\u003c/em\u003e rapidly increased during differentiation into plasma cells (\u003cb\u003eSupplementary Fig.\u0026nbsp;4-3h, i\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eEpithelial cell heterogeneity and their interactions with immune cells in the intestine\u003c/h2\u003e \u003cp\u003eWe analyzed 278,584 epithelial cells across41 tissues, identifying 50 subtypes (\u003cb\u003eSupplementary Fig.\u0026nbsp;5-1a and Supplementary Table\u0026nbsp;5\u0026thinsp;\u0026minus;\u0026thinsp;1\u003c/b\u003e). Although most subtypes were tissue-specific, keratinocytes (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e, \u003cspan citationid=\"CR114\" class=\"CitationRef\"\u003e114\u003c/span\u003e), spinous cells (22,672), basal cells (19,283), and goblet cells (13,055) exhibited the highest cell counts and were identified in more than six tissues (\u003cb\u003eSupplementary Fig.\u0026nbsp;5-1a\u003c/b\u003e). Cell types with similar biological functions were clustered together, such as chief cells, parietal cells, isthmus cells, mucous neck cells, and pit cells, all of which have the function of secreting gastric protease and promoting digestion (\u003cb\u003eSupplementary Fig.\u0026nbsp;5-1b\u003c/b\u003e). Additionally, three regions of the forestomach (rumen, reticulum, and omasum) and six regions of the intestine (duodenum, ileum, jejunum, colon, cecum, and rectum) were also clustered together (\u003cb\u003eSupplementary Fig.\u0026nbsp;5-1c\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eEpithelial cells play pivotal roles in absorption, metabolism, and immune defense\u003csup\u003e59,60\u003c/sup\u003e, making a thorough investigation of their functional and regulatory diversity essential, particularly within the digestive system. We scrutinized 156,832 epithelial cells across 15 digestive tissues, uncovering eight cell-to-tissue modules with distinct biological functions and TF regulation (\u003cb\u003eFig.\u0026nbsp;5a\u003c/b\u003e). For instance, epithelial cells in the rumen, reticulum, and omasum exhibited upregulated genes enriched in amino acid and fatty acid degradation, regulated by TFs such as OVOL2 and PITX1 (\u003cb\u003eFig.\u0026nbsp;5a\u003c/b\u003e). In comparison with the three ruminant forestomach compartments, the cell types of the human stomach and abomasum are nearly identical, and the TF regulation pattern of their epithelial cells is analogous. The highly expressed genes in the stomach and the abomasum are predominantly enriched in protein processing and amino acid processing, which is important for secreting gastric protease to facilitate digestion (\u003cb\u003eSupplementary Fig.\u0026nbsp;5-2a-d\u003c/b\u003e). In addition, the high-expression genes of the stomach were enriched in pyruvate and sugar metabolism, suggesting that the human stomach also has a similar function to the cattle forestomach (\u003cb\u003eSupplementary Fig.\u0026nbsp;5-2c\u003c/b\u003e). The developmental lineage of the forestomach and abomasum were clearly separated (\u003cb\u003eSupplementary Fig.\u0026nbsp;5-2e\u003c/b\u003e), supporting the multi-origin hypothesis that the forestomach originates from the esophagus while the abomasum originates from duodenum\u003csup\u003e61\u003c/sup\u003e. Genes involved in gastric acid secretion and protein processing were significantly activated during the development of the abomasum (\u003cb\u003eSupplementary Fig.\u0026nbsp;5-2f, g\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eTo explore the heterogeneity of spinous cells in the forestomach, we further divided them into 13 modules based on their gene co-expression patterns (\u003cb\u003eSupplementary Fig.\u0026nbsp;5-3a, b\u003c/b\u003e). Modules predominated in the rumen showing upregulated genes associated with fatty acid oxidation and proton transport (\u003cb\u003eSupplementary Fig.\u0026nbsp;5-3c, d\u003c/b\u003e), regulated by TFs including PITX1, HMGA1, HOXC6, and YBX1 (\u003cb\u003eSupplementary Fig.\u0026nbsp;5-3e\u003c/b\u003e). These findings underscore the crucial role of rumen spinous cells in fatty acid absorption. In addition, hepatocytes (HEPs) in the liver and bile duct exhibited upregulation of genes enriched in bile secretion and energy metabolism pathways, with specific TFs such as NR1H4 for bile salt synthesis, ONECUT2 for hepatocyte differentiation (\u003cb\u003eFig.\u0026nbsp;5a\u003c/b\u003e), and HNF4A and NFIA for fat, protein, and sugar metabolism (\u003cb\u003eSupplementary Fig.\u0026nbsp;5-4a-b\u003c/b\u003e). HEPs were further subdivided into three distinct subtypes (HEP0-HEP2; \u003cb\u003eSupplementary Fig.\u0026nbsp;5-4c\u003c/b\u003e), with HEP0 associated with bile secretion and VFA metabolism, HEP1 with glycolysis, and HEP2 with the calcium signaling pathway (\u003cb\u003eSupplementary Fig.\u0026nbsp;5-4d, e)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eEpithelial cells in the intestine were characterized by upregulated genes involved in fat and mineral digestion and absorption under the regulation of CDX1 and CDX2, two TFs implicated in intestinal cell differentiation and inflammation\u003csup\u003e62,63\u003c/sup\u003e (\u003cb\u003eFig.\u0026nbsp;5a\u003c/b\u003e). Given the pivotal role of goblet cells (GCs) in maintaining the intestinal immune barrier and inflammatory response\u003csup\u003e64\u003c/sup\u003e, we investigated cellular communication between GCs and seven types of immune cells. Interestingly, GCs had significantly stronger communication with conventional dendritic cells type 1 (cDC1) and macrophages in the ileum and cecum than in other intestinal tissues (\u003cb\u003eSupplementary Fig.\u0026nbsp;5-5a, b\u003c/b\u003e). Most of the ligand-receptor pairs were shared between GCs and immune cells, with the APP-CD74 pair showing the highest communication probability, in which CD74 signaling transduction is strongly activated during intestinal inflammation and protects the host by promoting epithelial cell regeneration, healing, and maintaining mucosal barrier integrity\u003csup\u003e65\u003c/sup\u003e (\u003cb\u003eFig.\u0026nbsp;5b-d\u003c/b\u003e). Significant cellular interactions among GCs were identified through CDH1-CDH1 pairs, which potentially maintain the intestinal barrier and have been shown to be important in the pathogenesis of ulcerative colitis and Crohn\u0026rsquo;s disease (CD)\u003csup\u003e66,67\u003c/sup\u003e. Moreover, GCs were further categorized into seven subtypes with distinct biological functions (GC0-GC6; \u003cb\u003eFig.\u0026nbsp;5e, f\u003c/b\u003e). For example, genes upregulated in GC1 were enriched in intracellular signal transduction, whereas those in GC4 were enriched in regulation of immune system processes, and exhibit the strongest cellular communication strength between GC1 and GC4 (\u003cb\u003eFig.\u0026nbsp;5f, g\u003c/b\u003e). Among the 128 identified significant cellular communications among GC subtypes and immune cells (\u003cb\u003eSupplementary Table\u0026nbsp;5\u0026thinsp;\u0026minus;\u0026thinsp;2\u003c/b\u003e), GC1 exhibited the strongest interactions with GC4 through the CDH signaling pathway, while GC5 and GC6 displayed no interplay with other GCs (\u003cb\u003eFig.\u0026nbsp;5h\u003c/b\u003e). GCs had stronger communication with macrophages (MA1) and cDC1 compared to other immune cells, predominantly via the APP signaling pathway, with GC1 and GC4 showing the strongest interactions (\u003cb\u003eFig.\u0026nbsp;5i\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eCellular basis and mechanisms underlying monogenic disorders in cattle\u003c/h2\u003e \u003cp\u003eTo explore whether the CattleCA could serve as a powerful resource for the dissection of the cellular basis and mechanisms underlying monogenic conditions in cattle, we compiled 183 causal genes associated with 145 bovine disorders from the Online Mendelian Inheritance in Animals (OMIA) database\u003csup\u003e68\u003c/sup\u003e and divided them into 10 trait domains based on their phenotypic manifestations (\u003cb\u003eSupplementary Table\u0026nbsp;6\u0026thinsp;\u0026minus;\u0026thinsp;1)\u003c/b\u003e. We detected 2,677 cell type-specific genes (z-score\u0026thinsp;\u0026gt;\u0026thinsp;0.75) across 129 distinct cell types (excluding PVALB\u0026thinsp;+\u0026thinsp;GABAergic neurons and proliferative cells) (\u003cb\u003eFig.\u0026nbsp;6a and Supplementary Table\u0026nbsp;6\u0026thinsp;\u0026minus;\u0026thinsp;2)\u003c/b\u003e. Our enrichment analysis revealed significant overlaps between cell lineage-specific genes and causal genes of disorders. For instance, muscle cell-specific genes were significantly enriched with muscle disorder genes (FDR\u0026thinsp;\u0026lt;\u0026thinsp;0.05), germline cells with reproduction disorders, and epithelial cells with skin disorders (\u003cb\u003eFig.\u0026nbsp;6b\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eNine out of 27 genes related to skin disorders exhibited significantly higher expression in epithelial cells compared to other cell lineages (\u003cb\u003eFig.\u0026nbsp;6c and Supplementary Fig.\u0026nbsp;6-1a\u003c/b\u003e). Disorder-related genes, such as \u003cem\u003eDSP\u003c/em\u003e for ichthyosis and \u003cem\u003eTSR2\u003c/em\u003e for hypotrichosis showed ubiquitous expression across various types of epithelial cells (\u003cb\u003eFig.\u0026nbsp;6d\u003c/b\u003e). \u003cem\u003eDSP\u003c/em\u003e, encoding the desmoplakin protein, plays a pivotal role in maintaining functional desmosomes crucial for skin cell integrity and adhesion\u003csup\u003e69\u003c/sup\u003e, while \u003cem\u003eTSR2\u003c/em\u003e is primarily involved in apoptosis\u003csup\u003e70\u003c/sup\u003e. Six other genes with pivotal roles in pigmentation, membrane production, and fat/enzyme transport, including \u003cem\u003eKRT5\u003c/em\u003e, \u003cem\u003eLAMA3\u003c/em\u003e, and \u003cem\u003eLAMC2\u003c/em\u003e for epidermolysis bullosa (EB), \u003cem\u003eABCA12\u003c/em\u003e for ichthyosis, \u003cem\u003eMLPH\u003c/em\u003e for coat color, and \u003cem\u003ePRLR\u003c/em\u003e for slick hair\u003csup\u003e71\u0026ndash;73\u003c/sup\u003e, displayed specific expression in duct and basal cells. Additionally, \u003cem\u003eSLC45A2\u003c/em\u003e, related to coat coloration, exhibited specific expression in melanocytes, and has previously been implicated in human melanin production\u003csup\u003e74\u003c/sup\u003e. To explore the cell-cell interaction of nine epithelial-specifically expressed disease genes, we performed cellular communication analysis, identifying laminin, encoded by \u003cem\u003eLAMA3\u003c/em\u003e and \u003cem\u003eLAMC2\u003c/em\u003e, as a key ligand in uninucleate trophoblast cells (UTCs) (found in the placenta) and progenitor cells (present in the jejunum and duodenum), respectively (\u003cb\u003eFig.\u0026nbsp;6e and Supplementary Fig.\u0026nbsp;6\u0026thinsp;\u0026minus;\u0026thinsp;2\u003c/b\u003e). \u003cem\u003eLAMA3\u003c/em\u003e showed high expression in placental uninucleate trophoblast cells, while \u003cem\u003eLAMC2\u003c/em\u003e exhibited elevated expression in intestinal progenitor cells (\u003cb\u003eFig.\u0026nbsp;6f and Supplementary Fig.\u0026nbsp;6-1b\u003c/b\u003e). \u003cem\u003eLAMA3\u003c/em\u003e and \u003cem\u003eLAMC2\u003c/em\u003e play crucial roles in regulating skin strength and resiliency\u003csup\u003e75\u003c/sup\u003e, with mutations disrupting laminin assembly and leading to EB in cattle\u003csup\u003e76\u003c/sup\u003e. A previous study has also reported that laminin deficiency can impact trophoblast differentiation and embryonic development\u003csup\u003e77\u003c/sup\u003e. Trajectory analysis of UTCs revealed specific expression of \u003cem\u003eLAMA3\u003c/em\u003e towards the end of UTC differentiation (\u003cb\u003eFig.\u0026nbsp;6g and Supplementary Fig.\u0026nbsp;6-1c\u003c/b\u003e). Furthermore, we annotated UTCs into seven subtypes (UTC0-UTC6; \u003cb\u003eSupplementary Fig.\u0026nbsp;6-1d\u003c/b\u003e). Among them, \u003cem\u003eLAMA3\u003c/em\u003e was predominantly expressed in the UTC2 and UTC3 (\u003cb\u003eFig.\u0026nbsp;6g and Supplementary Fig.\u0026nbsp;6-1c\u003c/b\u003e). Notably, UTC2 and UTC3, in which \u003cem\u003eLAMA3\u003c/em\u003e plays an important role as a ligand-encoding gene, exhibited stronger cellular communication with other cell types within the placenta compared to other five subtypes, indicating a potential correlation between EB and UTC2/UTC3 (\u003cb\u003eFig.\u0026nbsp;6h; Supplementary Fig.\u0026nbsp;6-1e; and Supplementary Fig.\u0026nbsp;6\u0026thinsp;\u0026minus;\u0026thinsp;3\u003c/b\u003e). Furthermore, marker genes of UTC2 and UTC3 were significantly enriched in several membrane pathways, including the basement membrane, the sarcolemma, and the apical plasma membrane (\u003cb\u003eSupplementary Fig.\u0026nbsp;6-1f\u003c/b\u003e). These pathways regulate cell-cell adhesion and affect the structure and stability of the basement membrane zone, which is the key pathogenic mechanism underlying EB\u003csup\u003e78\u003c/sup\u003e. In addition, \u003cem\u003eLAMA3\u003c/em\u003e, positioned within these pathways, provides a compelling hypothesis that \u003cem\u003eLAMA3\u003c/em\u003e, by participating in membrane function as a ligand-encoding gene within UTC2 and UTC3, may potentially contribute to EB in cattle (\u003cb\u003eSupplementary Table\u0026nbsp;6\u0026thinsp;\u0026minus;\u0026thinsp;3)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eFor muscle disorders, all 11 identified causal genes exhibited specific expression patterns (z score\u0026thinsp;\u0026gt;\u0026thinsp;0.75) within distinct muscle cells (\u003cb\u003eFig.\u0026nbsp;6i\u003c/b\u003e). Among them, \u003cem\u003eMYBPC1\u003c/em\u003e and \u003cem\u003ePPP1R13L\u003c/em\u003e displayed upregulated expression in skeletal muscle cells within the esophagus and cardiomyocytes in the heart, respectively (\u003cb\u003eFig.\u0026nbsp;6j and Supplementary Fig.\u0026nbsp;6-4a\u003c/b\u003e). \u003cem\u003eMYBPC1\u003c/em\u003e, identified as a candidate gene associated with increased muscular tonus (IMT) in both humans and cattle\u003csup\u003e79,80\u003c/sup\u003e, encodes the type I myonuclear isoform of myosin-binding protein C, an essential regulator of muscle contraction and tonus in humans\u003csup\u003e81\u003c/sup\u003e, therby underscoring a significant relationship between skeletal muscle cells and IMT. On the other hand, \u003cem\u003ePPP1R13L\u003c/em\u003e, encoding the inhibitor of apoptosis-stimulating p53 protein (iASPP), plays regulatory and inflammatory roles in cardiomyocytes, potentially inducing cardiomyopathy and woolly haircoat syndrome (CMWH) in dairy cattle\u003csup\u003e82\u003c/sup\u003e. We further classified skeletal muscle cells (SMCs) into five subtypes (SMC0-SMC4), and trajectory analysis revealed specific expression of \u003cem\u003eMYBPC1\u003c/em\u003e in SMC0 and at the late stage of skeletal muscle cell differentiation, corresponding to type I myonuclear cells (\u003cb\u003eFig.\u0026nbsp;6k; Supplementary Fig.\u0026nbsp;6-4b, c\u003c/b\u003e). Marker genes of SMC0 were predominantly enriched in muscle contraction, actin filament binding, and the Z disc \u003cb\u003e(Supplementary Fig.\u0026nbsp;6-4d\u003c/b\u003e), which were highly related to muscle tonus\u003csup\u003e83\u003c/sup\u003e. In addition, marker genes in SMC4 exhibited significant enrichment in immune response pathways, suggesting that skeletal muscle cells might have certain immune regulatory functions at the onset of differentiation (\u003cb\u003eFig.\u0026nbsp;6k and Supplementary Fig.\u0026nbsp;6-4d\u003c/b\u003e). Furthermore, we examined the expression patterns of four genes associated with reproductive disorders across four germline cell types and found that \u003cem\u003eABHD16B\u003c/em\u003e and \u003cem\u003eTMEM95\u003c/em\u003e exhibited upregulated expression in spermatids compared to other germline cells, associated with infertility and male subfertility, respectively (\u003cb\u003eSupplementary Fig.\u0026nbsp;6-4e\u003c/b\u003e). Additionally, specifically high expression of three blood/immune disorder genes, \u003cem\u003eITGB2\u003c/em\u003e, \u003cem\u003eF13A1\u003c/em\u003e, and \u003cem\u003eRASGRP2\u003c/em\u003e, was observed in immune cells (\u003cb\u003eSupplementary Fig.\u0026nbsp;6-4f, g\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eCellular basis and mechanisms underlying complex traits in cattle\u003c/h2\u003e \u003cp\u003eTo explore whether the CattleCA could contribute to unraveling some of the cellular basis and mechanisms underlying complex traits in cattle, we collected GWAS summary statistics for 55 complex traits, representing milk production (n\u0026thinsp;=\u0026thinsp;30), male fertility (n\u0026thinsp;=\u0026thinsp;5), coat color (n\u0026thinsp;=\u0026thinsp;6), immunoglobulin G (IgG) (n\u0026thinsp;=\u0026thinsp;10), body conformation (n\u0026thinsp;=\u0026thinsp;3) and health traits (n\u0026thinsp;=\u0026thinsp;1). To prioritize cell types involved in these complex traits, we conducted a trait-cell type enrichment analysis using scPagwas\u003csup\u003e84\u003c/sup\u003e, and revealed certain associations between cell lineages and complex traits. For instance, nerve cells were associated with milk production and germline cells with sperm traits (\u003cb\u003eFig.\u0026nbsp;7a and Supplementary Fig.\u0026nbsp;7\u0026thinsp;\u0026minus;\u0026thinsp;1\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eFor milk production traits, the most significantly associated cell types were neurons. Excitatory neurons, specifically, exhibited significant associations with milk fat yield (FY; \u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;8.25\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;19\u003c/sup\u003e) and capric acid (C10:0; \u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;1.51\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;20\u003c/sup\u003e), alongside amacrine cells displaying associations with lauric acid (C12:0; \u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;4.3\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;16\u003c/sup\u003e; \u003cb\u003eFig.\u0026nbsp;7b\u003c/b\u003e). This finding is consistent with our previous observation at the bulk tissue level in cattle, where a strong association between neurobiology and milk production traits was noted\u003csup\u003e85\u003c/sup\u003e, and aligns with previous research in humans suggesting reciprocal regulation between neuronal activity and lipid metabolism\u003csup\u003e86\u0026ndash;88\u003c/sup\u003e. Additionally, we observed that excitatory neurons in the cerebral cortex and amacrine cells in the retina exhibited significant associations and high trait-relevant scores (TRSs) for FY (\u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;8.87\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;6\u003c/sup\u003e), C10:0 (\u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;8.75\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;6\u003c/sup\u003e), and C12:0 (\u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;1.59\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e) as compared to other cell types, underscoring the important roles of these two tissues in regulating milk fat content (\u003cb\u003eFig.\u0026nbsp;7d and Supplementary Fig.\u0026nbsp;7-2a, b)\u003c/b\u003e. Furthermore, skeletal muscle cells showed a significant association with C12:0 (\u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;9.14\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;20\u003c/sup\u003e; \u003cb\u003eFig.\u0026nbsp;7b\u003c/b\u003e), consistent with their pivotal role in fatty acid utilization and intracellular fatty acid homeostasis\u003csup\u003e89,90\u003c/sup\u003e. Further analysis revealed that skeletal muscle cells in the esophagus (\u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;6.89\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;13\u003c/sup\u003e) and tongue (\u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;1.03\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;12\u003c/sup\u003e) showed significant associations with C12:0, indicating a prominent role for these two tissues in fatty acid regulation (\u003cb\u003eFig.\u0026nbsp;7e and Supplementary Fig.\u0026nbsp;7-2c\u003c/b\u003e). According to the pathway activity analysis conducted with scPagwas, neuronal excitement and lipolysis related pathways, like thermogenesis (\u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;1.15\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;198\u003c/sup\u003e), regulation of lipolysis in adipocytes (\u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;2.07\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;207\u003c/sup\u003e) and glutamatergic synapse (\u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;2\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;197\u003c/sup\u003e), had significant activity in excitatory neurons, amacrine cells and skeletal muscle cells respectively, and the genes in these pathways had significant enrichment with fatty acid trait-relevant genes (\u003cb\u003eFig.\u0026nbsp;7g, Supplementary Fig.\u0026nbsp;7-2f and Supplementary Table\u0026nbsp;7\u0026thinsp;\u0026minus;\u0026thinsp;1, 2\u003c/b\u003e). In addition, we observed that luminal cells, closely related to lactation and milk production traits\u003csup\u003e91\u003c/sup\u003e, exhibited a particularly significant association with the production of pentadecanoic acid (C15:0) (\u003cb\u003eFig.\u0026nbsp;7b\u003c/b\u003e), along with active involvement in pathways regulating lipolysis in adipocytes (\u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;2.11\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;218\u003c/sup\u003e), PI3K-Akt (\u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;1.73\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;292\u003c/sup\u003e), and retrograde endocannabinoid signaling pathways (\u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;3.2\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;286\u003c/sup\u003e)(\u003cb\u003eFig.\u0026nbsp;7g\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eFor male fertility traits, the most significant association was observed between sperm motilities (SMOT) and spermatocytes across 129 cell types (\u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;1.11\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;7\u003c/sup\u003e; \u003cb\u003eFig.\u0026nbsp;7c\u003c/b\u003e). Spermatocytes undergo a complex process of differentiation following meiosis, ultimately maturing into spermatids. This is a crucial step that directly impacts the quality of sperm produced\u003csup\u003e92\u003c/sup\u003e. This finding was further reinforced by the significant association observed between spermatocytes in the testis and SMOT (\u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;2.83\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;10\u003c/sup\u003e) within the testis tissue level, with the function of active pathways in spermatocytes primarily focused on nutrient metabolism (\u003cb\u003eFig.\u0026nbsp;7f, g\u003c/b\u003e). These pathways also had significant enrichment with SMOT-relevant genes of spermatocytes, indicating potential association of them with sperm production and maturation (\u003cb\u003eFig.\u0026nbsp;7g and Supplementary Table\u0026nbsp;7\u0026thinsp;\u0026minus;\u0026thinsp;2\u003c/b\u003e). Additionally, pDCs derived from PBMC were significantly associated with semen concentration per ejaculate (SCPE) trait (\u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;2.06\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;5\u003c/sup\u003e in the global atlas; \u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;7.31\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;4\u003c/sup\u003e in the PBMC; \u003cb\u003eFig.\u0026nbsp;7c and Supplementary Fig.\u0026nbsp;7-2e\u003c/b\u003e), consistent with our previous findings regarding the involvement of immune cells in male fertility traits\u003csup\u003e85\u003c/sup\u003e. This aligns with reports in humans indicating a significant association between DC abundance and sperm quality, suggesting a potential contribution of DC-mediated immune responses suboptimal to male fertility or infertility\u003csup\u003e93\u003c/sup\u003e. Pathway analysis revealed significant activity in distal convoluted tubule cells predominantly involving the regulation of hypothalamic gonadotropin-releasing hormone (GnRH) signaling pathway (\u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;8.83\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;250\u003c/sup\u003e; \u003cb\u003eFig.\u0026nbsp;7g\u003c/b\u003e). Furthermore, signal transduction pathways, like dopaminergic synapse, had significant activity in cone photoreceptor cells (\u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;=\u0026thinsp;5.59\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;265\u003c/sup\u003e; \u003cb\u003eFig.\u0026nbsp;7g\u003c/b\u003e). Additionally, the cell types most associated with coat color were type B intercalated cells, luminal cells, and amacrine cells; for IgG levels, they were type B intercalated cells, CD8\u003csup\u003e+\u003c/sup\u003e T cells, and skeletal muscle cells; and for body and health, they were chondrocytes, epithelial stem cells, and amacrine cells (\u003cb\u003eSupplementary Fig.\u0026nbsp;7-3a-c\u003c/b\u003e). In summary, the association of specific cell types with complex traits provides a cellular perspective for pinpointing the genetic regulatory mechanisms underlying important cattle traits.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eCross-species cell transcriptome similarity comparisons and disease associations\u003c/h2\u003e \u003cp\u003eTo investigate the similarities in gene expression, TF regulation, and cellular communication between cattle and humans at the single-cell level, we collected and analyzed 18 publicly available single-cell transcriptome datasets from 30 human tissues (\u003cb\u003eSupplementary Table\u0026nbsp;8\u0026thinsp;\u0026minus;\u0026thinsp;1\u003c/b\u003e). Using manual cell type annotation approach, we identified 106 distinct cell types, with 68 shared between cattle and humans. The Meta-neighbor analysis revealed conservation in gene expression between cattle and humans, particularly in nerve and immune cells (AUROC\u0026thinsp;\u0026gt;\u0026thinsp;0.90; \u003cb\u003eFig.\u0026nbsp;8a\u003c/b\u003e). Furthermore, the highly correlated cell-type regulon specificity scores (RSSs) of orthologous TFs (\u003cem\u003er\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.52\u0026ndash;0.77) also showed the evolutionary conservation of TF regulation, particularly in immune and epithelial cells (\u003cb\u003eFig.\u0026nbsp;8b, c; Supplementary Table\u0026nbsp;8\u0026thinsp;\u0026minus;\u0026thinsp;2 and Supplementary Fig.\u0026nbsp;8\u0026thinsp;\u0026minus;\u0026thinsp;1\u003c/b\u003e). For example, TFs crucial for B lymphocyte function, such as REL, ETS1, and ARID3A, exhibited RSS exceeding 0.25 in immune cells across both species, suggesting a conserved regulatory function in immune responses. Similarly, TFs such as HNF4G, CDX2, EHF, KLF4, and PITX1, implicated in epithelial cell differentiation and the establishment of the epithelial barrier\u003csup\u003e94\u0026ndash;98\u003c/sup\u003e, also had RSSs over 0.25 in both cattle and human epithelial cells, suggesting analogous regulatory mechanisms governing epithelial function. Furthermore, we observed notable similarities in cellular communication and signaling pathways, particularly in tissues like the small intestine and liver, between the two species (\u003cb\u003eFig.\u0026nbsp;8d-g and Supplementary Fig.\u0026nbsp;8\u003c/b\u003e). For instance, in the small intestine of the two species, enterocytes, goblet cells, Schwann cells, and B cells showed similar communication strengths and shared signaling pathways. In the liver, Kuffer cells, hepatocytes, and cholangiocytes demonstrated similar communication strength between the two species. However, several tissues, such as the mammary gland (\u003cb\u003eSupplementary Fig.\u0026nbsp;8\u0026thinsp;\u0026minus;\u0026thinsp;2\u003c/b\u003e), cerebellum (\u003cb\u003eSupplementary Fig.\u0026nbsp;8\u0026thinsp;\u0026minus;\u0026thinsp;2)\u003c/b\u003e, esophagus (\u003cb\u003eSupplementary Fig.\u0026nbsp;8\u0026thinsp;\u0026minus;\u0026thinsp;3)\u003c/b\u003e, and retina (\u003cb\u003eSupplementary Fig.\u0026nbsp;8\u0026thinsp;\u0026minus;\u0026thinsp;4)\u003c/b\u003e, displayed distinct cellular communication patterns between species. These findings together indicate considerable conservation at the cellular expression, TF regulation, and communication between cattle and humans, providing valuable insights into cross-species similarities.\u003c/p\u003e \u003cp\u003eTo explore whether the CattleCA resource can contribute to the explanation of genetic and cellular mechanisms underlying complex human traits and diseases, we analyzed the heritability enrichment of 43 human traits and diseases by linkage disequilibrium score regression (LDSC) on the orthologous marker genes across 49 cattle tissues. Significant enrichment of heritability for human complex traits and diseases was found in corresponding tissues in cattle (\u003cb\u003eSupplementary Table\u0026nbsp;8\u0026thinsp;\u0026minus;\u0026thinsp;3 and Supplementary Fig.\u0026nbsp;8-6a\u003c/b\u003e). For instance, the orthologous marker genes in cattle jejunum, colon, and reticulum showed significant enrichment of heritability for human inflammatory bowel disease (IBD) and CD, while markers in the blood, cecum, ileum, and pituitary gland showed enrichment for multiple sclerosis (MS) (FDR\u0026thinsp;\u0026lt;\u0026thinsp;0.05).\u003c/p\u003e \u003cp\u003eWe analyzed four diseases - MS, IBD, RA, and CD, due to their significant heritability enrichment and the potential links in their mechanisms of onset and progression in humans and cattle\u003csup\u003e99,100\u003c/sup\u003e. We calculated the heritability enrichment for these conditions across seven tissues by extending the categories of orthologous marker genes to cell types. Remarkably, certain cell types showed significant heritability enrichments, enhancing our understanding of their genetic and molecular basis. For instance, CD4\u003csup\u003e+\u003c/sup\u003e T cells exhibited substantial heritability enrichment for MS and RA, while CD8\u003csup\u003e+\u003c/sup\u003e T cells showed enrichment for CD and IBD (\u003cb\u003eFig.\u0026nbsp;8h-i\u003c/b\u003e; \u003cb\u003eSupplementary Fig.\u0026nbsp;8-6b-c and Supplementary Table\u0026nbsp;8\u0026thinsp;\u0026minus;\u0026thinsp;3\u003c/b\u003e). Moreover, several cell types displayed significant heritability enrichments for diseases have not been extensively studied. For example, microglia in the pituitary gland showed significant enrichment for MS (\u003cb\u003eFig.\u0026nbsp;8h\u003c/b\u003e), and the differentially expressed genes in pituitary microglia were enriched in signal transduction by p53 class mediator (\u003cb\u003eSupplementary Fig.\u0026nbsp;4-2j\u003c/b\u003e), suggesting that activation of the apoptotic protein p53 be associated with MS progression. Additionally, GCs in the colon were significantly enriched for IBD (\u003cb\u003eFig.\u0026nbsp;8i\u003c/b\u003e) and had strong communication with immune cells (\u003cb\u003eFig.\u0026nbsp;5b-d\u003c/b\u003e), suggesting their potential protective roles in the intestinal epithelial barrier.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn summary, through generating and analyzing gene expression of 1,793,854 cells across 59 tissue types in 15 cattle, we annotated 131 cell types to build the first version of CattleCA and provided a web portal for the community to explore and query all the results. The following comprehensive inter- and intra-tissue analyses advanced our understanding of cellular heterogeneity across the cattle body and between sexes. Integrating this resource with large scale population genetics data facilitated the detection of relevant cell types and stages for complex phenotypes. For instance, among all the annotated cell types, spermatocytes and excitatory neurons showed the strongest association with sperm motilities (SMOT) and milk fat yield, respectively, in dairy cattle. This knowledge can further sever as biological priors for prioritizing causal genes and variants as well as improving the prediction accuracy of complex phenotypes\u003csup\u003e85,101\u003c/sup\u003e. The cross-species comparative analysis revealed a high similarity in gene expression, regulation, and cellular communication between cattle and humans at single cell resolution. These findings will contribute to understanding the cellular and evolutionary mechanisms underlying zoonoses associated with cattle, such as tuberculosis, salmonellosis and ringworm\u003csup\u003e102\u003c/sup\u003e, facilitating the development of bovine models for certain human diseases\u003csup\u003e103,104\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eAlthough it provides a valuable resource for the cattle genetics and genomics community, the current CattleCA still has certain limitations: 1) Our current samples were limited to Holstein cattle, a dairy breed of global economic value. It is thus imperative to encompass a broader range of cattle breeds, such as beef cattle. 2) More biological and environmental contexts should be considered in the future development of CattleCA, such as embryonic stages, healthy status, and diet changes, because some cell types and states may only be observed in certain contexts. 3) Additional single-cell omics data, such as epigenome, proteome, and spatial transcriptome, are required for accurately and precisely annotating cell types and stages\u003csup\u003e105\u003c/sup\u003e. A more comprehensive CattleCA will enable a deeper understanding of the dynamic landscape of cellular function across diverse biological and environmental contexts, advancing our understanding of molecular mechanisms underlying complex phenotypes and environmental adaptations in cattle and even in humans.\u003c/p\u003e "},{"header":"Methods","content":"\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003cdiv id=\"Sec13\" class=\"Section3\"\u003e \u003ch2\u003eEthics statement\u003c/h2\u003e \u003cp\u003eThe experimental procedures were approved by the Institutional Animal Care and Use Committee (IACUC) at China Agricultural University (Beijing, China; approval number: DK996) and Northwest A\u0026amp;F University (Yangling, China; approval number: DK20230113), and the Animal Care Committee of Zhejiang University (Hangzhou, China; approval number ZJU202017326).\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eSample and data collection\u003c/h2\u003e \u003cp\u003eTwo healthy Chinese Holstein cows, sharing the same sire, similar birth and calving dates, and lactation performance, were selected from Beijing Sunlon Livestock Development Co., Ltd. (Beijing, China) were used. Both cows were in the mid to late stages of their first lactation. Following a 12-hr fasting period, animals were euthanized and 41 tissue samples were collected from each cow (see individuals 1 and 2 in \u003cb\u003eSupplementary Fig.\u0026nbsp;1\u0026ndash;1\u003c/b\u003e). The tissue samples were immediately placed into sterile serum-free Minimum Essential Medium (MEM) at 4\u0026deg;C and subsequently transported to the tissue culture laboratory for single-cell/nucleus suspension preparation within 4 hr. In addition, we also obtained single-cell/nucleus transcriptome data of Holstein cattle tissues from our collaborative teams\u003csup\u003e22,27\u003c/sup\u003e and public databases (rumen\u003csup\u003e21\u003c/sup\u003e, ovary\u003csup\u003e106\u003c/sup\u003e, placenta\u003csup\u003e107\u003c/sup\u003e, and intervertebral discs\u003csup\u003e108\u003c/sup\u003e). Overall, this study analyzed a total of 152 samples representing 59 tissues with one to nine replicates per tissue (including 10 tissues from different anatomical structures of tissues/organs, namely mammary parenchyma, mammary duct, mammary cistern, rumen papilla, rumen muscle, frontal lobe, occipital lobe, parietal lobe, circumvallate papilla, and fungal papilla) from 11 systems (comprising the digestive system, endocrine system, nervous system, cardiovascular system, respiratory system, urinary system, lymphatic system, female reproductive system, male reproductive system, skeletal system, and integumentary system) collected from 15 Holstein cattle (encompassing nine adult cows, one adult bull, one weaned female calf, three one-day-old female calf, and one female fetus) (\u003cb\u003eFig.\u0026nbsp;1b; Supplementary Fig.\u0026nbsp;1\u0026ndash;1 and Supplementary Table\u0026nbsp;1\u0026ndash;1\u003c/b\u003e). We also used single-cell transcriptome data from 30 human tissues for interspecies analysis (\u003cb\u003eSupplementary Table\u0026nbsp;8\u0026thinsp;\u0026minus;\u0026thinsp;1\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eSingle-cell suspension preparation\u003c/h2\u003e \u003cp\u003e \u003cb\u003eBone marrow\u003c/b\u003e: The fresh marrow form leg bone was minced and suspended i\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003en\u003c/span\u003e 5 mL of RPMI 1640 cell culture medium (Gibco). The cells were filtered through a 40 \u0026micro;m SmartStrainer (Miltenyi Biotec), centrifuged at 300 \u0026times;g for 10 min at 4℃, and resuspended in 1 mL of 1\u0026times; phosphate buffered saline (PBS). The cell suspension was mixed with Red Cell Lysis Solution (Solarbio) at a cell to lysis buffer ratio of 1:3, allowed to stand for 4\u0026ndash;8 min, centrifuged at 300 \u0026times;g for 5 min at 4℃, and resuspended in 1\u0026times;PBS with 0.04% bovine serum albumin (BSA).\u003c/p\u003e \u003cp\u003e \u003cb\u003eCervical lymph node and adrenal gland\u003c/b\u003e: The fresh tissues were washed with 1\u0026times;PBS, cut into 2 mm pieces, and incubated with dissociation enzymes according to manufacturer\u0026rsquo;s instructions (Tumor Dissociation Kit, Miltenyi Biotec) at 37℃ for 30\u0026ndash;60 min with gentle rotation. Then, the cell suspension was filtered through a 40 \u0026micro;m SmartStrainer, centrifuged at 300 \u0026times;g for 10 min at 4℃, and resuspended in 1 mL of 1\u0026times;PBS. The cell suspension was mixed with Red Cell Lysis Solution at a ratio of 1:3, allowed to stand for 3\u0026ndash;5 min, centrifuged at 300 \u0026times;g for 5 min at 4℃, and resuspended in 1\u0026times;PBS with 0.04% BSA.\u003c/p\u003e \u003cp\u003e \u003cb\u003eLiver, lung, and spleen\u003c/b\u003e: The fresh tissues were washed by 1\u0026times;PBS, cut into 2 mm pieces, and incubated with dissociation enzymes (Liver Dissociation Kit, Miltenyi Biotec) at 37℃ for 30\u0026ndash;60 min with gentle rotation. Then, the cell suspension was filtered through a 40 \u0026micro;m SmartStrainer, centrifuged at 300 \u0026times;g for 10 min at 4℃, and resuspended in 1 mL of 1\u0026times;PBS. The cell suspension was mixed with Red Cell Lysis Solution at a ratio of 1:3, allowed to stand for 3\u0026ndash;5 min, centrifuged at 300 \u0026times;g for 5 min at 4℃, and resuspended in 1\u0026times;PBS with 0.04% BSA.\u003c/p\u003e \u003cp\u003e \u003cb\u003eMammary parenchyma, mammary cistern, mammary duct, esophagus, sublingual gland, intima of uterine horn, intima of corpus uteri, rectum, jejunum, cecum, abomasum, duodenum, ileum, colon, rumen muscle and thymus\u003c/b\u003e: The fresh tissues were washed with 1\u0026times;PBS, cut into 2 mm pieces, and incubated with dissociation enzymes (Collagenase, Type II (Sigma), DNase I (Sigma-Aldrich), and 2% FBS (fetal bovine serum; Gibco)) at 37℃ for 20\u0026ndash;45 min with gentle rotation. Following cell dissociation, the cell suspension was filtered through a 70 \u0026micro;m SmartStrainer, centrifuged at 300 \u0026times;g for 10 min at 4℃, and resuspended in 1 mL of 1\u0026times;PBS. The cell suspension was mixed with Red Cell Lysis Solution at a ratio of 1:3, allowed to stand for 3\u0026ndash;5 min, centrifuged at 300 \u0026times;g for 5 min at 4℃, and resuspended in 1\u0026times;PBS with 0.04% BSA.\u003c/p\u003e \u003cp\u003e \u003cb\u003ePancreas\u003c/b\u003e: The fresh pancreas was washed by 1\u0026times;PBS, cut into 2 mm pieces, and incubated with dissociation enzymes (Collagenase, Type I (Sigma), DNase I, and 2% FBS) at 37℃ for 20\u0026ndash;60 min with gentle rotation. Then, the cell suspension was filtered through a 40 \u0026micro;m SmartStrainer, centrifuged at 300 \u0026times;g for 10 min at 4℃, and resuspended in 1 mL of 1\u0026times;PBS. The cell suspension was mixed with Red Cell Lysis Solution at a ratio of 1:3, allowed to stand for 3\u0026ndash;5 min, centrifuged at 300 \u0026times;g for 5 min at 4℃, and resuspended in 1\u0026times;PBS with 0.04% BSA.\u003c/p\u003e \u003cp\u003e \u003cb\u003ePBMC\u003c/b\u003e: Bovine peripheral blood (3 mL) was gently overlaid onto 3 mL of bovine peripheral blood lymphocyte separation solution (Solarbio) in a 15 mL conical centrifuge tube and centrifuged at 700\u0026times;g for 30 min at 20\u0026deg;C. After plasma was removed, the buffy coat containing PBMC was collected and transferred into a new 15 mL centrifuge tube containing 10 mL of 1\u0026times;PBS followed by centrifugation at 300 \u0026times;g for 10 min at 4℃. The PBMC was resuspended in 1 mL of 1\u0026times;PBS. The cell suspension was mixed with Red Cell Lysis Solution at a ratio of 1:3, allowed to stand for 3\u0026ndash;8 min, centrifuged at 300 \u0026times;g for 5 min at 4℃, and resuspended in 1\u0026times;PBS with 0.04% BSA.\u003c/p\u003e \u003cp\u003e \u003cb\u003ePituitary gland\u003c/b\u003e: The fresh pituitary gland was washed with 1\u0026times;PBS, cut into 2 mm pieces and incubated with dissociation enzymes (Adult Brain Dissociation Kit, Miltenyi Biotec) at 37℃ for 30\u0026ndash;60 min with gentle rotation. Then, the cell suspension was filtered through a 40 \u0026micro;m SmartStrainer and centrifuged at 300\u0026times;g for 10 min at 4℃. The cells were resuspended in 3.1 mL of 1\u0026times;PBS and mixed with 900 \u0026micro;L Debris Removal Solution (Miltenyi Biotec). The cell suspension was covered with 4 mL of pre-chilled 1\u0026times;PBS and centrifuged at 3000 \u0026times;g for 10 min at 4℃. The top and second layers of liquid were discarded, and 14 mL of pre-chilled 1\u0026times;PBS was added. The samples were centrifuged at 1000 \u0026times;g for 10 min at 4℃ and resuspended in 1 mL of 1\u0026times;PBS. The cell suspension was mixed with Red Cell Lysis Solution at a ratio of 1:3, allowed to stand for 3\u0026ndash;5 min, centrifuged at 300\u0026times;g for 5 min at 4℃, and resuspended in 1\u0026times;PBS with 0.04% BSA.\u003c/p\u003e \u003cp\u003e \u003cb\u003eReticulum, omasum, oviduct, circumvallate papilla, fungal papilla, and rumen papilla\u003c/b\u003e: The fresh tissues were washed by 1\u0026times;PBS, cut into 2 mm pieces, and incubated with dissociation enzymes (Collagenase, Type I, Collagenase, Type II, DNase I, and 2% FBS) at 37℃ for 30\u0026ndash;60 min with gentle rotation. Then, the cell suspension was filtered through a 40 \u0026micro;m SmartStrainer, centrifuged at 300 \u0026times;g for 10 min at 4℃, and resuspended in 1 mL of 1\u0026times;PBS. The cell suspension was mixed with Red Cell Lysis Solution at a ratio of 1:3, allowed to stand for 3\u0026ndash;5 min, centrifuged at 300 \u0026times;g for 5 min at 4℃, and resuspended in 1\u0026times;PBS with 0.04% BSA.\u003c/p\u003e \u003cp\u003e \u003cb\u003eSkin and ovary\u003c/b\u003e: The fresh tissues were washed by 1\u0026times;PBS, cut into 2 mm pieces, and incubated with dissociation enzymes (Collagenase, Type Ⅳ (Sigma), DNase I, and 2% FBS) at 37℃ for 30\u0026ndash;60 min with gentle rotation. Then, the cell suspension was filtered through a 40 \u0026micro;m SmartStrainer, centrifuged at 300 \u0026times;g for 10 min at 4℃, and resuspended in 1 mL of 1\u0026times;PBS. The cell suspension was mixed with Red Cell Lysis Solution at a ratio of 1:3, allowed to stand for 3\u0026ndash;5 min, centrifuged at 300 \u0026times;g for 5 min at 4℃, and resuspended in 1\u0026times;PBS with 0.04% BSA.\u003c/p\u003e \u003cp\u003e \u003cb\u003eThyroid\u003c/b\u003e: The fresh thyroid was washed by 1\u0026times;PBS, cut into 2 mm pieces, and incubated with dissociation enzymes (Multi Tissue Dissociation Kits 1, Miltenyi Biotec) at 37℃ for 20\u0026ndash;60 min with gentle rotation. Then, the cell suspension was filtered through a 70 \u0026micro;m SmartStrainer, centrifuged at 300 \u0026times;g for 5 min at 4℃, and resuspended in 1 mL of 1\u0026times;PBS with 0.04% BSA. The cell suspension was mixed with Red Cell Lysis Solution at a ratio of 1:3, allowed to stand for 3\u0026ndash;5 min, centrifuged at 300 \u0026times;g for 5 min at 4℃, and resuspended in 1\u0026times;PBS with 0.04% BSA.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eSingle-nucleus suspension preparation\u003c/h2\u003e \u003cp\u003e \u003cstrong\u003eLiver, Bile duct, heart, kidney, cerebral cortex, cerebellum, medulla oblongata, and hypothalamus\u003c/strong\u003e \u003cp\u003eThe single nuclei were isolated using the kit of Isolation of Nuclei for Single Cell RNA Sequencing (10\u0026times; Genomics). In short, tissues were placed in a 2 mL Eppendorf tube with 1 mL of pre-chilled LB buffer, minced with scissors, and incubated on ice for 1\u0026ndash;10 min. The tissue lysate was then filtered with a 40 \u0026micro;m SmartStrainer, centrifuged at 500 \u0026times;g for 5 min at 4℃, and resuspended with 300 \u0026micro;L of LB buffer. The nuclei were mixed with 300 \u0026micro;L of RB buffer. Then 600 \u0026micro;L PB1 and 600 \u0026micro;L PB2 buffer were sequentially underlaid from the very bottom of the tube. After centrifugation at 4000 \u0026times;g for 20 min at 4℃, the nuclei layer (150\u0026ndash;300 \u0026micro;L) was located at the junction of PB1 and PB2 solutions was then aspirated and mixed with 1 mL of RB buffer. The nuclei suspension was then filtered through a 40 \u0026micro;m SmartStrainer, centrifuged at 500 \u0026times;g for 5 min at 4℃, and resuspended in 100 \u0026micro;L of EB buffer.\u003c/p\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eLibrary construction and sequencing by 10\u0026times; Genomics platform\u003c/h2\u003e \u003cp\u003eThe cell/nucleus suspension (300\u0026ndash;600 living cells or nuclei/\u0026micro;L) was loaded onto the Chromium single cell controller (10\u0026times; Genomics) using Single Cell 3' Library and Gel Bead Kit V3.1 (10\u0026times; Genomics) and Chromium Single Cell G Chip Kit (10\u0026times; Genomics) to generate single-cell/nucleus gel beads in emulsion. Briefly, approximately 4,000 cells or nuclei were added to each channel, and the target cells/nuclei recovered were estimated to be about 2,000 cells/nuclei. Captured cells/nuclei were lysed and the released RNA was barcoded through reverse transcription in individual Gel Bead-In Emulsions (GEMs). Reverse transcription was performed on an S1000TM Touch Thermal Cycler (Bio-Rad) at 53\u0026deg;C for 45 min, followed by 85\u0026deg;C for 5 min and at 4\u0026deg;C. The cDNA was generated, amplified, and quality assessed using an Agilent 4200. The libraries were sequenced using an Illumina NovaSeq 6000 sequencer with a sequencing depth of at least 100,000 reads per cell with a pair-end 150 bp (PE150) strategy (CapitalBio Technology Co., Ltd., Beijing, China).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eRaw sequencing data processing\u003c/h2\u003e \u003cp\u003eThe \u003cem\u003eBos taurus\u003c/em\u003e ARS-UCD1.2 reference assembly\u003csup\u003e109\u003c/sup\u003e in FASTA format and annotated gene model in GTF format were downloaded from the Ensembl database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003eftp://ftp.ensembl.org/pub/release-101\u003c/span\u003e\u003cspan address=\"http://ftp://ftp.ensembl.org/pub/release-101\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The raw scRNA-seq/snRNA-seq data were aligned to the cattle reference genome and subjected to barcode assignment and unique molecular identifier (UMI) counting based on the Cell Ranger 7.0.1 pipeline (10\u003cb\u003e\u0026times;\u003c/b\u003e Genomics)\u003csup\u003e110\u003c/sup\u003e. For each library, ddqcR 0.1.0\u003csup\u003e111\u003c/sup\u003e was used to remove low-quality cells. It first clustered the cells using standard scRNA-seq analysis preprocessing and clustering steps. In each cluster, cells with n_counts and n_genes values lower than 2 median absolute deviations (MADs) from the median were filtered out. Additionally, cells with a mitochondrial genes percentage lower than 10% were removed. Doublets removal was performed using DoubletFinder 2.0.3\u003csup\u003e112\u003c/sup\u003e with the default parameter. It first averaged the transcriptional profile of randomly chosen cell pairs to create pseudo-doublets and then predicted doublets according to each real cell\u0026rsquo;s similarity in gene expression to the pseudo-doublets.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eCell clusters identification\u003c/h2\u003e \u003cp\u003eSeurat 4.0.6\u003csup\u003e113\u003c/sup\u003e was used to perform unsupervised clustering. Libraries from the same tissue were merged and underwent normalizing and scaling. Harmony 0.1.1\u003csup\u003e29\u003c/sup\u003e was used to correct four batch effects (sources, methods, platforms, and individuals) with the resetting parameters (Lambda\u0026thinsp;=\u0026thinsp;1, Theta\u0026thinsp;=\u0026thinsp;0.5). Variable genes were determined using Seurat\u0026rsquo;s Find-VariableGenes function with default parameters (selection.method = \u0026ldquo;vst\u0026rdquo;, nfeatures\u0026thinsp;=\u0026thinsp;2000). Clusters were identified via the FindClusters function (resolution\u0026thinsp;=\u0026thinsp;0.5) implemented in Seurat using the top 30 principal components and subsequently visualized using the RunUMAP function (reduction = \u0026ldquo;harmony\u0026rdquo;). Artificial annotation was performed on each cell cluster based on the marker genes reported in the relevant scientific literature.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eCell cycle index estimation\u003c/h2\u003e \u003cp\u003eTo obtain additional insights into the dynamic information about cell states, a cell cycle index was computed for each cell type using the CellCycleScoring function within Seurat\u003csup\u003e113\u003c/sup\u003e. The cells were categorized into G0/G1, S, and G2/M phases, denoting distinct cell cycle stages.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eCell type preference distribution analysis\u003c/h2\u003e \u003cp\u003eTo characterize the tissue distribution of mammary epithelial cells, odds ratios (ORs) were calculated and used to indicate preferences. Specifically, for each cell subtype i and tissue j, a 2 \u0026times; 2 matrix was constructed, which contained the number of cells of cell subtype i in tissue j, the number of cells of cell subtype i in other tissues, the number of cells of non-i cell subtype in tissue j, and the number of cells of non-i cell subtype in other tissues. Then Fisher\u0026rsquo;s exact test was applied to this matrix, thus, OR and the corresponding \u003cem\u003ep\u003c/em\u003e-value could be obtained. \u003cem\u003eP\u003c/em\u003e-values were adjusted using the Benjamini-Hochberg (BH) method implemented in the R function p.adjust. OR value\u0026thinsp;\u0026gt;\u0026thinsp;1.5 indicated that cell type i was preferentially distributed in tissue j, and an OR value\u0026thinsp;\u0026lt;\u0026thinsp;0.5 indicated that cell type i was not preferentially distributed in tissue.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003ePseudotime trajectory analysis\u003c/h2\u003e \u003cp\u003eMonocle2 2.26\u003csup\u003e114\u003c/sup\u003e was used to infer the state transition of the cell types/subtypes. The UMI count matrix of the cells was used to create the CellDataSet object and then filter out the genes expressed in fewer than 10 cells. The genes with qvalue\u0026thinsp;\u0026lt;\u0026thinsp;0.01 were identified as DEGs using the differentialGeneTest function and sorted by qvalue using the setOrderingFilter function. The pseudotime trajectory was constructed by the DDRTree algorithm using the default parameters. The dynamic expression changes of selected marker genes in pseudotime were visualised by the plot_genes_in_pseudotime and plot_pseudotime_heatmap functions. To explore the process of spermatogenesis, the germline cells were extracted from the testis and then re-descended for clustering with \u0026ldquo;pcs\u0026thinsp;=\u0026thinsp;20\u0026rdquo; and resolution set to 0.5. To explore the developmental process of cell types across tissues, cells were extracted and merged from each tissue, batch effects were corrected using Harmony 0.1.1 and then re-descended for clustering with \u0026ldquo;pcs\u0026thinsp;=\u0026thinsp;15\u0026rdquo; and resolution set to 0.1 for epithelial cells, 0.3 for B cells and 0.4 for myeloid cells. In order to shorten the model fitting time, cell subtypes with fewer than 2000 cells in immune cells were completely retained, while subtypes with more than 2000 cells were randomly sampled, and 25% of the cells were retained. To explore the mechanism of monogenic disorders genes in the process of cell differentiation, the uninuclear trophoblast cells and skeletal muscle cells were extracted from the placenta and esophagus, respectively, batch effects were corrected using Harmony 0.1.1 and then re-descended for clustering with \u0026ldquo;pcs\u0026thinsp;=\u0026thinsp;30\u0026rdquo; and resolution set to 0.1.\u003c/p\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003eCellular communication analysis\u003c/h2\u003e \u003cp\u003eCellular communication analyses were implemented using the CellChat 1.6.1 R package\u003csup\u003e38\u003c/sup\u003e for each tissue in label-based mode. Default parameters were used, except that min.cells was set to 10, which allows filtering out of cell types with the total number of cells less than 10. All the annotated cell types were classified based on their cell lineage. Interactions between different cell types were then aggregated, and the average intensity was computed to assess the comprehensive dynamics of cellular communication networks. Furthermore, cellular communication patterns within diverse tissues were examined in detail. For comparisons of cellular communication across tissues and species, the cellular communication analysis within specific tissues or species was first performed separately, and then the datasets were merged using mergeCellChat. Finally, the netVisual_diffInteraction function was used to compare and analyze the differences in communication strength.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003eGene regulatory network analysis\u003c/h2\u003e \u003cp\u003eGene regulatory network inference was performed using the Python package pySCENIC 0.12.1\u003csup\u003e39\u003c/sup\u003e with the default parameters. The raw counts, derived from the Seurat object, were based on one-to-one homologous genes between humans and cattle using Homologene 1.4.68.19.3.27. The human TF list (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://resources.aertslab.org/cistarget/tf_lists/allTFs_hg38.txt\u003c/span\u003e\u003cspan address=\"https://resources.aertslab.org/cistarget/tf_lists/allTFs_hg38.txt\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) was used as a reference for identifying co-expression modules through the GRNBoost2 algorithm. Then the regulons were obtained by detecting the genes directly targeted by the TF and removing other genes based on the enrichment of motifs within 10 kb from the target transcription start site (TSS) using cisTarget databases (Homo sapiens - hg38 - refseq_r80 - SCENIC\u0026thinsp;+\u0026thinsp;databases - Gene based (aertslab.org)). Using aucell, the regulon activity score (RAS) was measured as the area under the recovery curve. The activities associated with each cell type were evaluated by calculating the regulon specificity score (RSS)\u003csup\u003e115\u003c/sup\u003e. For cross-species cell-type-specific TFs analysis, the TFs datasets of dairy cattle and humans were separately inferred by pySCENIC. The Jensen Shannon divergence (JSD) was calculated according to the TF expressions, and the TF-specific score was defined as 1-\u0026radic;JSD. The z-score was then calculated to normalize the TF-specific score to predict the basic TF in each cell type by the following formula: \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(z score=\\frac{{x}_{ij}-{\\mu }_{i}}{{\\sigma }_{i}}\\)\u003c/span\u003e\u003c/span\u003e; Where \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({x}_{ij}\\)\u003c/span\u003e\u003c/span\u003e was the RSS for TF j in cell type i, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({\\mu }_{i}\\)\u003c/span\u003e\u003c/span\u003e was the average RSS of all the TFs in cell type i, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({\\sigma }_{i}\\)\u003c/span\u003e\u003c/span\u003e was the standard deviation for RSS of all TFs in cell type i. In total, 500 cells of each cell type were selected for cross-species analyses. The correlation coefficient \u003cem\u003er\u003c/em\u003e and \u003cem\u003ep\u003c/em\u003e-value were calculated by corr.test (method = \u0026ldquo;pearson\u0026rdquo;, adjust = \u0026ldquo;fdr\u0026rdquo;).\u003c/p\u003e \u003cdiv id=\"Sec25\" class=\"Section3\"\u003e \u003ch2\u003eTranscription factor conservative analysis\u003c/h2\u003e \u003cp\u003eFor each cell type within a given tissue, if a regulon is positive in over 25% of cells, the TF is then considered \u0026ldquo;positive\u0026rdquo; in that specific cell type. PhastCons\u003csup\u003e116\u003c/sup\u003e and phyloP\u003csup\u003e117\u003c/sup\u003e conservation scores calculated on multiple sequence alignment of sequences of 100 species of vertebrates were retrieved using UCSC table browser data retrieval tool\u003csup\u003e118\u003c/sup\u003e (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://hgdownload.soe.ucsc.edu/goldenPath/hg38/\u003c/span\u003e\u003cspan address=\"https://hgdownload.soe.ucsc.edu/goldenPath/hg38/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). TF binding site data in cattle were classified into four groups based on the regulation of cell types, and a Wilcoxon test was conducted to evaluate the significance of differences in TF regulation among these groups.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section3\"\u003e \u003ch2\u003eSex bias classification of tissue analysis\u003c/h2\u003e \u003cp\u003eThe 19 tissues were classified into three categories based on sex bias in cell type composition: non-sex-biased regulation, partial, and sex-biased regulation. Non-sex-biased tissues had no sex-biased cell clusters, partial tissues had less than 50% sex-biased clusters, and tissues with over 50% sex-biased clusters were labeled as sex-biased. The classification was established utilizing the paired Wilcoxon test based on RAS, with subsequent adjustment of \u003cem\u003ep\u003c/em\u003e-values using the Bonferroni method for multiple comparisons.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section3\"\u003e \u003ch2\u003eCell type-specific transcription factor network construction\u003c/h2\u003e \u003cp\u003eFor each cell type within a given tissue, a transcription factor (TF) is deemed \u0026ldquo;positive\u0026rdquo; in that particular cell type if its regulon is positive in more than 25% of cells, and it also exhibits a z-score value exceeding 0.75.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec28\" class=\"Section2\"\u003e \u003ch2\u003eFunction enrichment analysis\u003c/h2\u003e \u003cp\u003eGene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were performed using the clusterProfiler 4.0\u003csup\u003e119\u003c/sup\u003e. The GO terms and KEGG pathways of selected genes were enriched in the org.Bt.eg.db and bta databases, using the erichGO and enrichKEGG functions, respectively, with a threshold parameter of \u0026ldquo;pvalueCutoff\u0026thinsp;=\u0026thinsp;0.05\u0026rdquo;. GSVA 1.49.6\u003csup\u003e120\u003c/sup\u003e was used to investigate functional differences between cell subtypes. The bovine KEGG database was downloaded from MsigDB using the msigdbr 7.5.1 (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/igordot/msigdbr\u003c/span\u003e\u003cspan address=\"https://github.com/igordot/msigdbr\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), after which the gsva function was used to calculate the pathway scores in the pathway for each subtype-specific gene set.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec29\" class=\"Section2\"\u003e \u003ch2\u003eCell type diversity analysis\u003c/h2\u003e \u003cp\u003eThe Shannon entropy\u003csup\u003e121\u003c/sup\u003e was calculated to evaluate cell type diversity in each tissue according to the formula \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(-{\\sum }_{\\text{i}}({\\text{p}}_{\\text{i}}\\times {\\text{l}\\text{o}\\text{g}}_{2}({\\text{p}}_{\\text{i}}\\left)\\right)\\)\u003c/span\u003e\u003c/span\u003e, where \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({\\text{p}}_{\\text{i}}\\)\u003c/span\u003e\u003c/span\u003e is the proportion of cell type in cell class i for each tissue. They were then plotted in R 4.2.0 using ggplot2 3.4.1 and complexHeatmap 2.15.4\u003csup\u003e122\u003c/sup\u003e .\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eCell type conservation analysis\u003c/h3\u003e\n\u003cp\u003eMetaNeighbor 1.18\u003csup\u003e123\u003c/sup\u003e was used to provide a measure of the replicability of cell types within and across species. The highly variable genes were identified using the variableGenes function and the correlations between cell types were determined based on AUROC values using MetaNeighborUS analysis. In the cross-species analysis, 500 cells of each cell type were selected.\u003c/p\u003e \u003cdiv id=\"Sec31\" class=\"Section2\"\u003e \u003ch2\u003eCell function scoring\u003c/h2\u003e \u003cp\u003eTo evaluate the extracellular and intracellular antigen presentation abilities of myeloid cells and B cells, as well as macrophage subpopulations of the M1 and M2 phenotypes, the AddModuleScore function in the Seurat package was used to calculate the antigen presentation score (APS) and macrophage_type score. Using the \u0026ldquo;special\u0026thinsp;=\u0026thinsp;Bos taurus, category\u0026thinsp;=\u0026thinsp;C5, subcategory\u0026thinsp;=\u0026thinsp;GO: BP\u0026rdquo; parameter of the msigdbr package (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/igordot/msigdbr\u003c/span\u003e\u003cspan address=\"https://github.com/igordot/msigdbr\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) to obtain gene set (\u003cb\u003eSupplementary Table\u0026nbsp;4\u0026thinsp;\u0026minus;\u0026thinsp;2\u003c/b\u003e) from \u0026ldquo;GOBP_ANTIGEN_PROCESSING_AND_PRESENTATION_OF_EXOGENOUS_PEPTIDE_ANTIGEN_VIA_MHC_CLASS_I\u0026rdquo;, \u0026ldquo;GOBP_ANTIGEN_PROCESSING_AND_PRESENTATION_OF_PEPTIDE_ANTIGEN_VIA_MHC_CLASS_I\u0026rdquo;, \u0026ldquo;GOBP_ANTIGEN_PROCESSING_AND_PRESENTATION_OF_EXOGENOUS_PEPTIDE_ANTIGEN_VIA_MHC_CLASS_II\u0026rdquo;, \u0026ldquo;GOBP_ANTIGEN_PROCESSING_AND_PRESENTATION_OF_PEPTIDE_OR_POLYSACCHARIDE_ANTIGEN_VIA_MHC_CLASS_II\u0026rdquo; in the msigdb database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.gsea-msigdb.org/gsea/msigdb\u003c/span\u003e\u003cspan address=\"https://www.gsea-msigdb.org/gsea/msigdb\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The classically activated M1 macrophages gene sets are \u003cem\u003eSOCS1\u003c/em\u003e, \u003cem\u003eNOS2\u003c/em\u003e, \u003cem\u003eTNF\u003c/em\u003e, \u003cem\u003eCXCL9\u003c/em\u003e, \u003cem\u003eCXCL10, CXCL11\u003c/em\u003e, \u003cem\u003eCD86\u003c/em\u003e, \u003cem\u003eIL1A\u003c/em\u003e, \u003cem\u003eIL1B\u003c/em\u003e, \u003cem\u003eIL6\u003c/em\u003e, \u003cem\u003eCCL5\u003c/em\u003e, \u003cem\u003eIRF5\u003c/em\u003e, \u003cem\u003eIRF1\u003c/em\u003e, and \u003cem\u003eCCR7\u003c/em\u003e, while the selectively activated M2 macrophages gene sets are \u003cem\u003eIL4R\u003c/em\u003e, \u003cem\u003eCCL4\u003c/em\u003e, \u003cem\u003eCCL18\u003c/em\u003e, \u003cem\u003eCCL22\u003c/em\u003e, \u003cem\u003eMARCO\u003c/em\u003e, \u003cem\u003eVEGFA\u003c/em\u003e, \u003cem\u003eCTSA\u003c/em\u003e, \u003cem\u003eCTSB\u003c/em\u003e, \u003cem\u003eTGFB1\u003c/em\u003e, \u003cem\u003eMMP9\u003c/em\u003e, \u003cem\u003eCLEC7A\u003c/em\u003e, \u003cem\u003eMSR1\u003c/em\u003e, \u003cem\u003eIRF4\u003c/em\u003e, \u003cem\u003eCD163\u003c/em\u003e, \u003cem\u003eTGM2\u003c/em\u003e, and \u003cem\u003eMRC1\u003c/em\u003e. Data were compared by the two-tailed Student\u0026rsquo;s \u003cem\u003et\u003c/em\u003e-test (**** \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.0001; ns p\u0026thinsp;\u0026gt;\u0026thinsp;0.05).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec32\" class=\"Section2\"\u003e \u003ch2\u003eHigh-dimensional weighted correlation network analysis\u003c/h2\u003e \u003cp\u003eThe hdWGCNA 0.2.24\u003csup\u003e124\u003c/sup\u003e was used to perform high-dimensional weighted gene co-expression network analysis based on single-cell data. First, we input the genes expressed in at least 5% of the cells and used the MetacellsByGroups function to construct the metacell gene expression matrix. Then, the TestSoftPowers function was used to determine the soft power. The ConstructNetwork function was used to build the co-expression network. All analyses were conducted according to standard procedures (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://smorabit.github.io/hdWGCNA/articles/hdWGCNA.html\u003c/span\u003e\u003cspan address=\"https://smorabit.github.io/hdWGCNA/articles/hdWGCNA.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e \u003cdiv id=\"Sec33\" class=\"Section3\"\u003e \u003ch2\u003eCattle monogenic disorder genes collection\u003c/h2\u003e \u003cp\u003eA total of 634 monogenic disorders in cattle were assembled from the Online Mendelian Inheritance in Animals (OMIA) database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.omia.org/home\u003c/span\u003e\u003cspan address=\"https://www.omia.org/home\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Candidate genes related to 145 disorders were identified. To facilitate systematic analysis, these disorders were classified into 10 groups based on their phenotypic manifestations, encompassing conditions related to blood and immune systems, connective tissue, embryonic development, embryonic lethality, metabolic processes, muscle function, neural disorders, reproductive issues, skin-related disorders, and tissue-specific conditions (\u003cb\u003eSupplementary Table\u0026nbsp;6\u0026thinsp;\u0026minus;\u0026thinsp;1\u003c/b\u003e). In order to ensure the robustness of the statistical analyses, a minimum threshold of more than five candidate genes was imposed for each disorder group.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec34\" class=\"Section3\"\u003e \u003ch2\u003eCell-type specific gene identification\u003c/h2\u003e \u003cp\u003eCell-type specific genes were identified using the z-score factor. Initially, the count matrix underwent a transformation into the Counts Per Million (CPM) values, and the average was computed within each cell type to create a pseudo-bulk expression matrix\u003csup\u003e17\u003c/sup\u003e. Subsequently, tspex 0.6.2\u003csup\u003e125\u003c/sup\u003e was utilized to calculate z-score values under the log10-transformed CPM matrix. Genes exhibiting a z-score value\u0026thinsp;\u0026gt;\u0026thinsp;0.75 in each cell type were designated as cell-type-specific genes and cell-type-specific genes in a given cell lineage were defined as cell lineage-specific genes.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e\n\u003ch3\u003eEnrichment analysis between cell types and monogenic disorders\u003c/h3\u003e\n\u003cp\u003eThe correlation between cell-type-specific genes and monogenic disorder genes was assessed through Chi-square and Fisher\u0026rsquo;s tests, utilizing a 2\u0026times;2 matrix consisting of intersected genes, disorder genes, cell-type-specific genes, and all genes. The Chi-square test was employed when expected frequencies for all elements exceeded 5; otherwise, Fisher\u0026rsquo;s exact test was used. Subsequently, the false discovery rate (FDR) values were computed using the Bonferroni method to address multiple comparisons. Additionally, ORs were also calculated to validate the accuracy of significant results based on the same 2\u0026times;2 matrix.\u003c/p\u003e\n\u003ch3\u003eGene coding DNA sequence (CDS) region alignment between humans and cattle\u003c/h3\u003e\n\u003cp\u003eDNA sequences of the gene CDS region were downloaded from the NCBI-Genbank (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ncbi.nlm.nih.gov/genbank/\u003c/span\u003e\u003cspan address=\"https://www.ncbi.nlm.nih.gov/genbank/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Sequence alignment was performed based on the NCBI-Blast method (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://blast.ncbi.nlm.nih.gov/Blast.cgi\u003c/span\u003e\u003cspan address=\"https://blast.ncbi.nlm.nih.gov/Blast.cgi\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e \u003cdiv id=\"Sec37\" class=\"Section2\"\u003e \u003ch2\u003eGenome-wide association study (GWAS) analysis\u003c/h2\u003e \u003cp\u003eThe SNP BeadChip and phenotypic data of 16,188 Chinese Holstein cows were assembled, including 9,045 bovine 150K BeadChip, 1,505 bovine 80K BeadChip, and 5,638 bovine 50K BeadChip, from the Dairy Association of China. The phenotypic data included 55 traits, spanning milk production, sperm, coat color, lgG, body conformation, and health. All SNP BeadChip data were mapped to the bovine reference genome (ARS-UCD1.2) and 132,961 SNPs were obtained after imputing to the 150K level using Beagle 5.1\u003csup\u003e126\u003c/sup\u003e. They were then imputed again to genome-scale sequence level using a high-quality sequencing imputation panel of 28,166,177 SNPs based on 3,530 cattle. After filtering out low-quality SNPs with \u0026ldquo;Dosage R-Squared (DR2)\u0026thinsp;\u0026lt;\u0026thinsp;0.9, Minor Allele Frequency (MAF)\u0026thinsp;\u0026lt;\u0026thinsp;0.05 and Hardy-Weinberg Equilibrium (HWE) test result \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.0001\u0026rdquo;, a total of 8,535,460 SNPs were obtained. Finally, GWAS analysis was performed using GCTA 1.94.0\u003csup\u003e127\u003c/sup\u003e with the \u0026ldquo;--fastGWA-mlm\u0026rdquo; option, and the results were visualized using CMplot 4.4.1\u003csup\u003e128\u003c/sup\u003e.\u003c/p\u003e \u003cdiv id=\"Sec38\" class=\"Section3\"\u003e \u003ch2\u003eEnrichment analysis between cell types and complex traits\u003c/h2\u003e \u003cp\u003eScpagwas 1.3.0\u003csup\u003e84\u003c/sup\u003e was utilized to perform the enrichment analysis between cell types and complex traits. It employs a polygenic regression model to prioritize a set of trait-relevant genes and uncover trait-relevant cell subpopulations by incorporating pathway activity transformed scRNA-seq data with GWAS summary data. To enhance the comprehensiveness of our results, 319 human KEGG pathways downloaded from the KEGG database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.genome.jp/kegg\u003c/span\u003e\u003cspan address=\"https://www.genome.jp/kegg\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) were used after eliminating duplicates and converting homologous genes. The \u003cem\u003eBoot_evaluate\u003c/em\u003e function was employed to identify the significant trait-relevant relevant cell types and calculate trait-relevant scores. The \u003cem\u003escGet_PCC\u003c/em\u003e function was used to prioritize the top trait-relevant genes by ranking the Pearson correlation coefficient (PCC). Genes with the top 50 PCC values were defined as trait-relevant genes in each cell type. In addition, the \u003cem\u003escPagwas_perform_score\u003c/em\u003e function was applied to perform pathway activity analysis and define the significance of active pathways in each cell type based on the singular value decomposition (SVD) method. Enrichment analysis between trait-relevant genes and active pathway genes was performed based on Fisher\u0026rsquo;s test and Chi-square test within each cell type.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec39\" class=\"Section2\"\u003e \u003ch2\u003eLinkage disequilibrium score regression (LDSC)\u003c/h2\u003e \u003cp\u003eLDSC\u003csup\u003e129\u003c/sup\u003e was to detect whether the heritability of a phenotype is enriched around highly specifically expressed genes in a given tissue or cell type. All SNPs associated with the trait were obtained from publicly available data (\u003cb\u003eSupplementary Table\u0026nbsp;8\u0026thinsp;\u0026minus;\u0026thinsp;3\u003c/b\u003e). The comparison of tissue contributions involved the selection of the top 200 DEGs (log\u003csub\u003e2\u003c/sub\u003e FC\u0026thinsp;\u0026ge;\u0026thinsp;1.5 and FDR\u0026thinsp;\u0026le;\u0026thinsp;0.05), which were sorted by FDR from least to most for each tissue as a category. All DEG categories for tissues were collectively inputted to run LDSC for traits. Subsequently, the \u003cem\u003ep\u003c/em\u003e-values were computed using the BH method to account for multiple comparisons. Similarly, the top 200 DEGs of each cell type cluster meeting the same criteria were selected for analysis, and the results were visualized using ggplot2 3.4.1.\u003c/p\u003e \u003cdiv id=\"Sec40\" class=\"Section3\"\u003e \u003ch2\u003eStatistical and reproducibility\u003c/h2\u003e \u003cp\u003eNo statistical method was used to predetermine the sample size. The details of data exclusions for each specific analysis are available in the \u003cspan refid=\"Sec12\" class=\"InternalRef\"\u003eMethods\u003c/span\u003e section. The experiments were not randomized, as all the datasets are publicly available from observational studies. The investigators were not blinded to allocation during experiments and outcome assessment, as the data were not from controlled randomized studies.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003ch2\u003eData and code availability\u003c/h2\u003e\n\u003cp\u003eAll single-cell and single-nucleus RNA sequencing data newly generated in this study are available for download under accession number PRJNA1119173 from SRA (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ncbi.nlm.nih.gov/sra/\u003c/span\u003e\u003cspan address=\"https://www.ncbi.nlm.nih.gov/sra/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) \u003cem\u003evia\u003c/em\u003e the provided reviewer link (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://dataview.ncbi.nlm.nih.gov/object/PRJNA1119173?reviewer=po13jm18jl62b78mi9k7r1mju7/\u003c/span\u003e\u003cspan address=\"https://dataview.ncbi.nlm.nih.gov/object/PRJNA1119173?reviewer=po13jm18jl62b78mi9k7r1mju7/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The processed datasets and expression profiles of annotated cell types are available via the CattleCA web portal: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://ngdc.cncb.ac.cn/cattleca/\u003c/span\u003e\u003cspan address=\"https://ngdc.cncb.ac.cn/cattleca/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. All the computational codes are freely available \u003cem\u003evia\u003c/em\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/FarmGTEx/CattleCellAtlas_pipeline_V0/\u003c/span\u003e\u003cspan address=\"https://github.com/FarmGTEx/CattleCellAtlas_pipeline_V0/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/p\u003e\n\u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eThis work was financially supported by the National Key R\u0026amp;D Program of China (2021YFF1000700, 2022YFF1000103); the National Natural Science Foundation of China (32372836); STI 2030-Major Projects (2023ZD04069); and the Program for Changjiang Scholar and Innovation Research Team in University (IRT_15R62). Lingzhao Fang was supported from Agriculture and Food Research Initiative Competitive grants nos. 2022-67015-36215 (H.Z.) from the USDA National Institute of Food and Agriculture. Jayne C. Hope was funded by the Biotechnology and Biological Sciences Research Council through Institute Strategic Programme Funding (grant number BBS/E/RL/230002B). J.F.O. is supported by the Science Foundation Ireland (SFI) Centre for Research Training in Genomics Data Science (grant no. 18/CRT/6214). Bingjie Li acknowledges funding from the UK Biotechnology and Biological Sciences Research Council (BBSRC) with grant BB/X009505/1. We acknowledge the support of the High-Performance Computing Platform of China Agricultural University (Beijing) and the Xihe High-Performance Computing Platform of the National Research Facility for Phenotypic and Genotypic Analysis of Model Animals (Beijing).\u003c/p\u003e\n\u003ch2\u003eAuthor contributions\u003c/h2\u003e\n\u003cp\u003eAll authors made substantial contributions to the conception or design of the study; the acquisition, analysis or interpretation of data; or drafting or revising the paper. D. Sun, L. Fang, B. Han, H. Sun, Y. Jiang, and G. E. Liu conceived and designed the project. D. Sun, Y. Jiang, H. Sun, Z. Ma, and L. Liu provided samples and data. H. Li, B. Han, S. Zhu, and T. Shi performed bioinformatic analyses of single cell/nucleus RNA-seq data. Q. Zhang, A. Chen, Y. Song, W. Ye, A. Du, Y. Fu, M. Jia, and T. Shi annotated the cell types manually. Q. Zhang and B. Han conducted the intra-tissue cellular heterogeneity. W. Zheng and B. Han conducted the inter-tissue cellular heterogeneity. H. Li and W. Zheng conducted integrative analysis with genetic variants. A. Chen and B. Han conducted comparative analysis between cattle and humans. Y. Hou, Z. Zhang, D. Zou, and Z. Yuan built the CattleCA web portal. D. Sun, L. Fang, and B. Han contributed to the data and computational resources. D. Sun, L. Fang, G. E. Liu, Y. Hou, F. Wang, H. Sun, Y. Jiang, W. Liu, W. Tuo, and J. C. Hope contributed to the critical interpretation of analytical results before and during manuscript preparation. B. Han, H. Li, Q. Zhang, W. Zheng, and A. Chen drafted the manuscript. L. Fang, D. Sun, G. E. Liu, Y. Hou, F. Wang, H. Sun, Y. Jiang, W. Liu, W. Tuo, J. C. Hope, D. E. MacHugh, J. F. O. Grady, O. Madsen, G. Sahana, Y. Luo, L. Lin, C. Li, Z. Cai, B. Li, and Z. Zhang revised the manuscript. All authors read, edited and approved the final manuscript.\u003c/p\u003e\n\u003ch2\u003eCompeting interests\u003c/h2\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eBruford, M.W., Bradley, D.G. \u0026amp; Luikart, G. DNA markers reveal the complexity of livestock domestication. Nat Rev Genet 4, 900\u0026ndash;10 (2003).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAjmone-Marsan, P., Garcia, J.F. \u0026amp; Lenstra, J.A. On the origin of cattle: How aurochs became cattle and colonized the world. 19, 148\u0026ndash;157 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePitt, D. \u003cem\u003eet al.\u003c/em\u003e Domestication of cattle: Two or three events? Evol Appl 12, 123\u0026ndash;136 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRitter, C., Beaver, A. \u0026amp; von Keyserlingk, M.A.G. The complex relationship between welfare and reproduction in cattle. Reproduction in Domestic Animals 54, 29\u0026ndash;37 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLaborde, D., Mamun, A., Martin, W., Pi\u0026ntilde;eiro, V. \u0026amp; Vos, R. Agricultural subsidies and global greenhouse gas emissions. Nature Communications 12(2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu, X.M. \u003cem\u003eet al.\u003c/em\u003e Global greenhouse gas emissions from animal-based foods are twice those of plant-based foods. Nature Food 2, 724-+ (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHawkins, J.W. \u003cem\u003eet al.\u003c/em\u003e High-yield dairy cattle breeds improve farmer incomes, curtail greenhouse gas emissions and reduce dairy import dependency in Tanzania. Nature Food 3, 957-+ (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu, Z.-L., Park, C.A. \u0026amp; Reecy, J.M. Bringing the Animal QTLdb and CorrDB into the future: meeting new challenges and providing updated services. Nucleic Acids Research 50, D956-D961 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFreebern, E. \u003cem\u003eet al.\u003c/em\u003e GWAS and fine-mapping of livability and six disease traits in Holstein cattle. BMC Genomics 21, 41 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHosseinzadeh, S., Rafat, S.A., Javanmard, A. \u0026amp; Fang, L. Identification of candidate genes associated with milk production and mastitis based on transcriptome-wide association study. Anim Genet 55, 430\u0026ndash;439 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJiang, J. \u003cem\u003eet al.\u003c/em\u003e Functional annotation and Bayesian fine-mapping reveals candidate genes for important agronomic traits in Holstein bulls. Commun Biol 2, 212 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJiang, L. \u003cem\u003eet al.\u003c/em\u003e Genome wide association studies for milk production traits in Chinese Holstein population. PLoS One 5, e13661 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, S. \u003cem\u003eet al.\u003c/em\u003e A multi-tissue atlas of regulatory variants in cattle. Nat Genet 54, 1438\u0026ndash;1447 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHan, X. \u003cem\u003eet al.\u003c/em\u003e Construction of a human cell landscape at single-cell level. Nature 581, 303\u0026ndash;309 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTabula Sapiens, C. \u003cem\u003eet al.\u003c/em\u003e The Tabula Sapiens: A multiple-organ, single-cell transcriptomic atlas of humans. Science 376, eabl4896 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHan, X. \u003cem\u003eet al.\u003c/em\u003e Mapping the Mouse Cell Atlas by Microwell-Seq. Cell 173, 1307 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, H. \u003cem\u003eet al.\u003c/em\u003e Fly Cell Atlas: A single-nucleus transcriptomic atlas of the adult fruit fly. Science 375, eabk2432 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWagner, D.E. \u003cem\u003eet al.\u003c/em\u003e Single-cell mapping of gene expression landscapes and lineage in the zebrafish embryo. Science 360, 981\u0026ndash;987 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYe, F. \u003cem\u003eet al.\u003c/em\u003e Construction of the axolotl cell landscape using combinatorial hybridization sequencing at single-cell resolution. Nat Commun 13, 4228 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, F. \u003cem\u003eet al.\u003c/em\u003e Endothelial cell heterogeneity and microglia regulons revealed by a pig cell landscape at single-cell level. Nat Commun 13, 3620 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGao, Y. \u003cem\u003eet al.\u003c/em\u003e Single-cell transcriptomic analyses of dairy cattle ruminal epithelial cells during weaning. Genomics 113, 2045\u0026ndash;2055 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu, J.J. \u003cem\u003eet al.\u003c/em\u003e Microbiota-host crosstalk in the newborn and adult rumen at single-cell resolution. BMC Biol 20, 280 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXue, M.Y. \u003cem\u003eet al.\u003c/em\u003e Investigation of fiber utilization in the rumen of dairy cows based on metagenome-assembled genomes and single-cell RNA sequencing. Microbiome 10, 11 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGao, Y. \u003cem\u003eet al.\u003c/em\u003e Single-cell transcriptomic and chromatin accessibility analyses of dairy cattle peripheral blood mononuclear cells and their responses to lipopolysaccharide. BMC Genomics 23, 338 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGu, F. \u003cem\u003eet al.\u003c/em\u003e Gut microbiome is linked to functions of peripheral immune cells in transition cows during excessive lipolysis. Microbiome 11, 40 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, L. \u003cem\u003eet al.\u003c/em\u003e A single-cell atlas of bovine skeletal muscle reveals mechanisms regulating intramuscular adipogenesis and fibrogenesis. J Cachexia Sarcopenia Muscle 14, 2152\u0026ndash;2167 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu, J.J. \u003cem\u003eet al.\u003c/em\u003e Cross-tissue single-cell transcriptomic landscape reveals the key cell subtypes and their potential roles in the nutrient absorption and metabolism in dairy cattle. J Adv Res 37, 1\u0026ndash;18 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu, J.J. \u003cem\u003eet al.\u003c/em\u003e Age- and Microbiota-Dependent Cell Stemness Plasticity Revealed by Cattle Cell Landscape. Research (Wash D C) 6, 0025 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKorsunsky, I. \u003cem\u003eet al.\u003c/em\u003e Fast, sensitive and accurate integration of single-cell data with Harmony. Nat Methods 16, 1289\u0026ndash;1296 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStrandberg, Y. \u003cem\u003eet al.\u003c/em\u003e Lipopolysaccharide and lipoteichoic acid induce different innate immune responses in bovine mammary epithelial cells. Cytokine 31, 72\u0026ndash;86 (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorato, A., Martignani, E., Miretti, S., Baratta, M. \u0026amp; Accornero, P. External and internal EGFR-activating signals drive mammary epithelial cells proliferation and viability. Molecular and Cellular Endocrinology 520(2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMipam, T. \u003cem\u003eet al.\u003c/em\u003e Single-cell transcriptome analysis and in vitro differentiation of testicular cells reveal novel insights into male sterility of the interspecific hybrid cattle-yak. Bmc Genomics 24(2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuo, J.T. \u003cem\u003eet al.\u003c/em\u003e The adult human testis transcriptional cell atlas. Cell Research 28, 1141\u0026ndash;1157 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu, S., Hu, Y.C., Liu, H.F. \u0026amp; Shi, Y. Loss of YY1 Impacts the Heterochromatic State and Meiotic Double-Strand Breaks during Mouse Spermatogenesis. Molecular and Cellular Biology 29, 6245\u0026ndash;6256 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWawrzik, M., Spiess, A.N., Herrmann, R., Buiting, K. \u0026amp; Horsthemke, B. Expression of SNURF\u0026ndash;SNRPN upstream transcripts and epigenetic regulatory genes during human spermatogenesis. European Journal of Human Genetics 17, 1463\u0026ndash;1470 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFimia, G.M., De Cesare, D. \u0026amp; Sassone-Corsi, P. CBP-independent activation of CREM and CREB by the LIM-only protein ACT. Nature 398, 165\u0026ndash;169 (1999).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBechtel, T.J., Reyes-Robles, T., Fadeyi, O.O. \u0026amp; Oslund, R.C. Strategies for monitoring cell-cell interactions. Nat Chem Biol 17, 641\u0026ndash;652 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJin, S.Q. \u003cem\u003eet al.\u003c/em\u003e Inference and analysis of cell-cell communication using CellChat. Nature Communications 12(2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAibar, S. \u003cem\u003eet al.\u003c/em\u003e SCENIC: single-cell regulatory network inference and clustering. Nature Methods 14, 1083-+ (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFischer, C. \u003cem\u003eet al.\u003c/em\u003e AGMO Inhibitor Reduces 3T3-L1 Adipogenesis. Cells 10(2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHelsley, R.N. \u003cem\u003eet al.\u003c/em\u003e Ketohexokinase-C regulates global protein acetylation to decrease carnitine palmitoyltransferase 1a-mediated fatty acid oxidation. Journal of Hepatology 79(2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, S.Y. \u003cem\u003eet al.\u003c/em\u003e Loss of hepatic FTCD promotes lipid accumulation and hepatocarcinogenesis by upregulating PPARc and SREBP2. Jhep Reports 5(2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, T. \u003cem\u003eet al.\u003c/em\u003e Hippocalcin-Like 1 blunts liver lipid metabolism to suppress tumorigenesis via directly targeting RUVBL1-mTOR signaling. Theranostics 12, 7450\u0026ndash;7464 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcLatchie, L.M. \u003cem\u003eet al.\u003c/em\u003e RAMPs regulate the transport and ligand specificity of the calcitonin-receptor-like receptor. Nature 393, 333\u0026ndash;339 (1998).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStutchfield, B.M. \u0026amp; Forbes, S.J. Liver sinusoidal endothelial cells in disease\u0026ndash;and for therapy? J Hepatol 58, 178\u0026ndash;80 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHammoutene, A. \u0026amp; Rautou, P.E. Role of liver sinusoidal endothelial cells in non-alcoholic fatty liver disease. J Hepatol 70, 1278\u0026ndash;1291 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVyas, J.M., Van der Veen, A.G. \u0026amp; Ploegh, H.L. The known unknowns of antigen processing and presentation. Nature Reviews Immunology 8, 607\u0026ndash;618 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePishesha, N., Harmand, T.J. \u0026amp; Ploegh, H.L. A guide to antigen processing and presentation. Nature Reviews Immunology 22, 751\u0026ndash;764 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMass, E., Nimmerjahn, F., Kierdorf, K. \u0026amp; Schlitzer, A. Tissue-specific macrophages: how they develop and choreograph tissue biology. Nature Reviews Immunology (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHashimoto, D. \u003cem\u003eet al.\u003c/em\u003e Tissue-Resident Macrophages Self-Maintain Locally throughout Adult Life with Minimal Contribution from Circulating Monocytes. Immunity 38, 792\u0026ndash;804 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDavies, L.C., Jenkins, S.J., Allen, J.E. \u0026amp; Taylor, P.R. Tissue-resident macrophages. Nature Immunology 14, 986\u0026ndash;995 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchuijs, M.J., Hammad, H. \u0026amp; Lambrecht, B.N. Professional and 'Amateur' Antigen-Presenting Cells In Type 2 Immunity. Trends in Immunology 40, 22\u0026ndash;34 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoche, P.A. \u0026amp; Furuta, K. The ins and outs of MHC class II-mediated antigen processing and presentation. Nature Reviews Immunology 15, 203\u0026ndash;216 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGreen, M.R. \u003cem\u003eet al.\u003c/em\u003e Signatures of murine B-cell development implicate Yy1 as a regulator of the germinal center-specific program. Proc Natl Acad Sci U S A 108, 2873\u0026ndash;8 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang, C., Geng, H., Boss, I., Wang, L. \u0026amp; Melnick, A. Cooperative transcriptional repression by BCL6 and BACH2 in germinal center B-cell differentiation. Blood 123, 1012\u0026ndash;20 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCousu, C. \u003cem\u003eet al.\u003c/em\u003e Germinal center output is sustained by HELLS-dependent DNA-methylation-maintenance in B cells. Nat Commun 14, 5695 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCalame, K.L., Lin, K.I. \u0026amp; Tunyaplin, C. Regulatory mechanisms that determine the development and function of plasma cells. Annu Rev Immunol 21, 205\u0026ndash;30 (2003).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHe, S. \u003cem\u003eet al.\u003c/em\u003e Single-cell transcriptome profiling of an adult human cell atlas of 15 major organs. Genome Biol 21, 294 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEngevik, A.C., Kaji, I. \u0026amp; Goldenring, J.R. The Physiology of the Gastric Parietal Cell. Physiol Rev 100, 573\u0026ndash;602 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGroschwitz, K.R. \u0026amp; Hogan, S.P. Intestinal barrier function: molecular regulation and disease pathogenesis. J Allergy Clin Immunol 124, 3\u0026ndash;20; quiz 21\u0026thinsp;\u0026ndash;\u0026thinsp;2 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, L. \u003cem\u003eet al.\u003c/em\u003e Large-scale ruminant genome sequencing provides insights into their evolution and distinct traits. Science 364(2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSoubeyran, P. \u003cem\u003eet al.\u003c/em\u003e Cdx1 promotes differentiation in a rat intestinal epithelial cell line. Gastroenterology 117, 1326\u0026ndash;38 (1999).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSilberg, D.G., Swain, G.P., Suh, E.R. \u0026amp; Traber, P.G. Cdx1 and cdx2 expression during intestinal development. \u003cem\u003eGastroenterology\u003c/em\u003e 119, 961\u0026thinsp;\u0026ndash;\u0026thinsp;71 (2000).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGustafsson, J.K. \u0026amp; Johansson, M.E.V. The role of goblet cells and mucus in intestinal homeostasis. Nat Rev Gastroenterol Hepatol 19, 785\u0026ndash;803 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFarr, L. \u003cem\u003eet al.\u003c/em\u003e CD74 Signaling Links Inflammation to Intestinal Epithelial Cell Regeneration and Promotes Mucosal Healing. Cell Mol Gastroenterol Hepatol 10, 101\u0026ndash;112 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLees, C.W., Barrett, J.C., Parkes, M. \u0026amp; Satsangi, J. New IBD genetics: common pathways with other diseases. Gut 60, 1739\u0026ndash;53 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMuise, A.M. \u003cem\u003eet al.\u003c/em\u003e Polymorphisms in E-cadherin (CDH1) result in a mis-localised cytoplasmic protein that is associated with Crohn's disease. Gut 58, 1121\u0026ndash;7 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNicholas, F.W. Online Mendelian Inheritance in Animals (OMIA): a record of advances in animal genetics, freely available on the Internet for 25 years. Anim Genet 52, 3\u0026ndash;9 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBolling, M.C. \u003cem\u003eet al.\u003c/em\u003e Lethal acantholytic epidermolysis bullosa due to a novel homozygous deletion in DSP: expanding the phenotype and implications for desmoplakin function in skin and heart. Br J Dermatol 162, 1388\u0026ndash;94 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHe, H.J., Bing, H. \u0026amp; Liu, G. TSR2 Induces laryngeal cancer cell apoptosis through inhibiting NF-kappaB signaling pathway. Laryngoscope 128, E130-E134 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAkiyama, M. The roles of ABCA12 in keratinocyte differentiation and lipid barrier formation in the epidermis. \u003cem\u003eDermatoendocrinol\u003c/em\u003e 3, 107\u0026thinsp;\u0026ndash;\u0026thinsp;12 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBetz, R.C. \u003cem\u003eet al.\u003c/em\u003e Loss-of-function mutations in the keratin 5 gene lead to Dowling-Degos disease. Am J Hum Genet 78, 510\u0026ndash;9 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFukuda, M., Kuroda, T.S. \u0026amp; Mikoshiba, K. Slac2-a/melanophilin, the missing link between Rab27 and myosin Va: implications of a tripartite protein complex for melanosome transport. J Biol Chem 277, 12432\u0026ndash;6 (2002).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGraf, J., Hodgson, R. \u0026amp; van Daal, A. Single nucleotide polymorphisms in the MATP gene are associated with normal human pigmentation variation. Hum Mutat 25, 278\u0026ndash;84 (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAumailley, M. \u003cem\u003eet al.\u003c/em\u003e A simplified laminin nomenclature. Matrix Biol 24, 326\u0026ndash;32 (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKiritsi, D., Has, C. \u0026amp; Bruckner-Tuderman, L. Laminin 332 in junctional epidermolysis bullosa. Cell Adh Migr 7, 135\u0026ndash;41 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, M., Yin, Y., Yu, H. \u0026amp; Zhou, R. Laminins Regulate Placentation and Pre-eclampsia: Focus on Trophoblasts and Endothelial Cells. Front Cell Dev Biol 8, 754 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUitto, J. \u0026amp; Richard, G. Progress in epidermolysis bullosa: genetic classification and clinical implications. Am J Med Genet C Semin Med Genet 131C, 61\u0026ndash;74 (2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWiedemar, N., Riedi, A.K., Jagannathan, V., Dr\u0026ouml;gem\u0026uuml;ller, C. \u0026amp; Meylan, M. Genetic Abnormalities in a Calf with Congenital Increased Muscular Tonus. Journal of Veterinary Internal Medicine 29, 1418\u0026ndash;1421 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShashi, V. \u003cem\u003eet al.\u003c/em\u003e Heterozygous variants in MYBPC1 are associated with an expanded neuromuscular phenotype beyond arthrogryposis. Human Mutation 40, 1115\u0026ndash;1126 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWeber, F.E., Vaughan, K.T., Reinach, F.C. \u0026amp; Fischman, D.A. Complete sequence of human fast-type and slow-type muscle myosin-binding-protein C (MyBP-C). Differential expression, conserved domain structure and chromosome assignment. Eur J Biochem 216, 661\u0026ndash;9 (1993).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSimpson, M.A. \u003cem\u003eet al.\u003c/em\u003e A mutation in NFkappaB interacting protein 1 causes cardiomyopathy and woolly haircoat syndrome of Poll Hereford cattle. Anim Genet 40, 42\u0026ndash;6 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGanguly, J., Kulshreshtha, D., Almotiri, M. \u0026amp; Jog, M. Muscle Tone Physiology and Abnormalities. Toxins (Basel) 13(2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMa, Y. \u003cem\u003eet al.\u003c/em\u003e Polygenic regression uncovers trait-relevant cellular contexts through pathway activation transformation of single-cell RNA sequencing data. Cell Genom 3, 100383 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFang, L.Z. \u003cem\u003eet al.\u003c/em\u003e Comprehensive analyses of 723 transcriptomes enhance genetic and biological interpretations for complex traits in cattle. Genome Research 30, 790\u0026ndash;801 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, M. \u003cem\u003eet al.\u003c/em\u003e Gut-brain circuits for fat preference. Nature 610, 722\u0026ndash;730 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLinehan, V., Fang, L.Z. \u0026amp; Hirasawa, M. Short-term high-fat diet primes excitatory synapses for long-term depression in orexin neurons. J Physiol 596, 305\u0026ndash;316 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWallace, C.W. \u0026amp; Fordahl, S.C. Obesity and dietary fat influence dopamine neurotransmission: exploring the convergence of metabolic state, physiological stress, and inflammation on dopaminergic control of food intake. Nutr Res Rev 35, 236\u0026ndash;251 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSacchetti, M., Saltin, B., Osada, T. \u0026amp; van Hall, G. Intramuscular fatty acid metabolism in contracting and non-contracting human skeletal muscle. J Physiol 540, 387\u0026ndash;95 (2002).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWatt, M.J. \u0026amp; Hoy, A.J. Lipid metabolism in skeletal muscle: generation of adaptive and maladaptive intracellular signals for cellular function. Am J Physiol Endocrinol Metab 302, E1315-28 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTwigger, A.J. \u003cem\u003eet al.\u003c/em\u003e Transcriptional changes in the mammary gland during lactation revealed by single cell sequencing of cells from human milk. Nat Commun 13, 562 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNeto, F.T., Bach, P.V., Najari, B.B., Li, P.S. \u0026amp; Goldstein, M. Spermatogenesis in humans and its affecting factors. Semin Cell Dev Biol 59, 10\u0026ndash;26 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDuan, Y.G. \u003cem\u003eet al.\u003c/em\u003e Dendritic cells in semen of infertile men: association with sperm quality and inflammatory status of the epididymis. \u003cem\u003eFertil Steril\u003c/em\u003e 101, 70\u0026ndash;77 e3 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, L. \u003cem\u003eet al.\u003c/em\u003e A reinforcing HNF4\u0026ndash;SMAD4 feed-forward module stabilizes enterocyte identity. 51, 777\u0026ndash;785 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGao, N., White, P. \u0026amp; Kaestner, K.H. Establishment of intestinal identity and epithelial-mesenchymal signaling by Cdx2. Developmental cell 16, 588\u0026ndash;599 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTugores, A. \u003cem\u003eet al.\u003c/em\u003e The epithelium-specific ETS protein EHF/ESE-3 is a context-dependent transcriptional repressor downstream of MAPK signaling cascades. Journal of Biological Chemistry 276, 20397\u0026ndash;20406 (2001).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFlandez, M., Guilmeau, S., Blache, P. \u0026amp; Augenlicht, L. KLF4 regulation in intestinal epithelial cell maturation. Experimental cell research 314, 3712\u0026ndash;3723 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRouco, R. \u003cem\u003eet al.\u003c/em\u003e Cell-specific alterations in Pitx1 regulatory landscape activation caused by the loss of a single enhancer. 12, 7235 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNaser, S.A., Sagramsingh, S.R., Naser, A.S. \u0026amp; Thanigachalam, S.J.W.J.o.G.W. Mycobacterium avium subspecies paratuberculosis causes Crohn's disease in some inflammatory bowel disease patients. 20, 7403 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBendele, A. \u003cem\u003eet al.\u003c/em\u003e Animal models of arthritis: relevance to human disease. 27, 134\u0026ndash;142 (1999).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHamel, A.R. \u003cem\u003eet al.\u003c/em\u003e Integrating genetic regulation and single-cell expression with GWAS prioritizes causal genes and cell types for glaucoma. Nature Communications 15(2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcDaniel, C.J., Cardwell, D.M., Moeller, R.B. \u0026amp; Gray, G.C. Humans and Cattle: A Review of Bovine Zoonoses. Vector-Borne and Zoonotic Diseases 14, 1\u0026ndash;19 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSaied, A.A., Metwally, A.A., Mohamed, H.M.A. \u0026amp; Haridy, M.A.M. The contribution of bovines to human health against viral infections. Environmental Science and Pollution Research 28, 46999\u0026ndash;47023 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHamernik, D.L. Farm animals are important biomedical models. Animal Frontiers 9, 3\u0026ndash;5 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZeng, H.K. What is a cell type and how to define it? Cell 185, 2739\u0026ndash;2755 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSoto, D.A. \u0026amp; Ross, P.J. Similarities between bovine and human germline development revealed by single-cell RNA sequencing. Reproduction 161, 239\u0026ndash;253 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDavenport, K.M. \u003cem\u003eet al.\u003c/em\u003e Single-nuclei RNA sequencing (snRNA-seq) uncovers trophoblast cell types and lineages in the mature bovine placenta. Proc Natl Acad Sci U S A 120, e2221526120 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePanebianco, C.J., Dave, A., Charytonowicz, D., Sebra, R. \u0026amp; Iatridis, J.C. Single-cell RNA-sequencing atlas of bovine caudal intervertebral discs: Discovery of heterogeneous cell populations with distinct roles in homeostasis. FASEB J 35, e21919 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRosen, B.D. \u003cem\u003eet al.\u003c/em\u003e De novo assembly of the cattle reference genome with single-molecule sequencing. Gigascience 9(2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZheng, G.X. \u003cem\u003eet al.\u003c/em\u003e Massively parallel digital transcriptional profiling of single cells. Nat Commun 8, 14049 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSubramanian, A., Alperovich, M., Yang, Y. \u0026amp; Li, B. Biology-inspired data-driven quality control for scientific discovery in single-cell transcriptomics. Genome Biol 23, 267 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcGinnis, C.S., Murrow, L.M. \u0026amp; Gartner, Z.J. DoubletFinder: Doublet Detection in Single-Cell RNA Sequencing Data Using Artificial Nearest Neighbors. \u003cem\u003eCell Syst\u003c/em\u003e 8, 329\u0026ndash;337 e4 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHao, Y. \u003cem\u003eet al.\u003c/em\u003e Integrated analysis of multimodal single-cell data. Cell 184, 3573\u0026ndash;3587 e29 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTrapnell, C. \u003cem\u003eet al.\u003c/em\u003e The dynamics and regulators of cell fate decisions are revealed by pseudotemporal ordering of single cells. Nature Biotechnology 32, 381\u0026ndash;386 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSuo, S.B. \u003cem\u003eet al.\u003c/em\u003e Revealing the Critical Regulators of Cell Identity in the Mouse Cell Atlas. Cell Reports 25, 1436-+ (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSiepel, A. \u003cem\u003eet al.\u003c/em\u003e Evolutionarily conserved elements in vertebrate, insect, worm, and yeast genomes. Genome Research 15, 1034\u0026ndash;1050 (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePollard, K.S., Hubisz, M.J., Rosenbloom, K.R. \u0026amp; Siepel, A. Detection of nonneutral substitution rates on mammalian phylogenies. Genome Research 20, 110\u0026ndash;121 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKarolchik, D. \u003cem\u003eet al.\u003c/em\u003e The UCSC Table Browser data retrieval tool. Nucleic Acids Research 32, D493-D496 (2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu, T. \u003cem\u003eet al.\u003c/em\u003e clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation (Camb) 2, 100141 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eH\u0026auml;nzelmann, S., Castelo, R. \u0026amp; Guinney, J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics 14, 7 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEraslan, G. \u003cem\u003eet al.\u003c/em\u003e Single-nucleus cross-tissue molecular reference maps toward understanding disease gene function. Science 376, eabl4290 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGu, Z., Eils, R. \u0026amp; Schlesner, M. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics 32, 2847\u0026ndash;9 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCrow, M., Paul, A., Ballouz, S., Huang, Z.J. \u0026amp; Gillis, J. Characterizing the replicability of cell types defined by single cell RNA-sequencing data using MetaNeighbor. Nature Communications 9(2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorabito, S. \u003cem\u003eet al.\u003c/em\u003e Single-nucleus chromatin accessibility and transcriptomic characterization of Alzheimer's disease. Nat Genet 53, 1143\u0026ndash;1155 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCamargo, A.P., Vasconcelos, A.A., Fiamenghi, M.B., Pereira, G.A.G. \u0026amp; Carazzolle, M.F. tspex: a tissue-specificity calculator for gene expression data. (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrowning, B.L., Zhou, Y. \u0026amp; Browning, S.R. A One-Penny Imputed Genome from Next-Generation Reference Panels. Am J Hum Genet 103, 338\u0026ndash;348 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJiang, L., Zheng, Z., Fang, H. \u0026amp; Yang, J. A generalized linear mixed model association tool for biobank-scale data. Nat Genet 53, 1616\u0026ndash;1621 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYin, L. \u003cem\u003eet al.\u003c/em\u003e rMVP: A Memory-efficient, Visualization-enhanced, and Parallel-accelerated Tool for Genome-wide Association Study. Genomics Proteomics Bioinformatics 19, 619\u0026ndash;628 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBulik-Sullivan, B.K. \u003cem\u003eet al.\u003c/em\u003e LD Score regression distinguishes confounding from polygenicity in genome-wide association studies. Nature genetics 47, 291\u0026ndash;295 (2015).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-4631710/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4631710/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSystematic characterization of the molecular states of cells in livestock tissues is essential for understanding cellular and genetic mechanisms underlying economically and ecologically important physiological traits. This knowledge contributes to the advancement of sustainable and precision agriculture-food systems. Here, as part of the Farm animal Genotype-Tissue Expression (FarmGTEx) project, we describe a comprehensive reference map comprising 1,793,854 cells from 59 bovine tissues, spanning both sexes and multiple developmental stages. This map, generated by single-cell/nucleus RNA sequencing, identifies 131 distinct cell types, revealing intra- and inter-tissue cellular heterogeneity in gene expression, transcription factor regulation, and intercellular communication. Integrative analysis with genetic variants that underpin bovine monogenic and complex traits uncovers cell types of relevance, such as spermatocytes responsible for sperm motilities and excitatory neurons for milk fat yield. Comparative analysis reveals similarities in gene expression between cattle and humans at single-cell resolution, allowing for detection of relevant cell types for studying human complex phenotypes. This cattle cell atlas will serve as a key resource for cattle genetics and genomics, immunology, comparative biology, and ultimately human biomedicine.\u003c/p\u003e","manuscriptTitle":"Cattle Cell Atlas: a multi-tissue single cell expression repository for advanced bovine genomics and comparative biology","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-07-03 06:26:10","doi":"10.21203/rs.3.rs-4631710/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-genetics","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"ng","sideBox":"Learn more about [Nature Genetics](http://www.nature.com/ng/)","snPcode":"","submissionUrl":"","title":"Nature Genetics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Research","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"e996dc70-f458-4c32-bc55-71374977a4fb","owner":[],"postedDate":"July 3rd, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":34009196,"name":"Biological sciences/Genetics/Genomics/Transcriptomics"},{"id":34009197,"name":"Biological sciences/Computational biology and bioinformatics/Data processing"},{"id":34009198,"name":"Biological sciences/Molecular biology/Transcriptomics"}],"tags":[],"updatedAt":"2025-09-06T07:09:14+00:00","versionOfRecord":{"articleIdentity":"rs-4631710","link":"https://doi.org/10.1038/s41588-025-02329-5","journal":{"identity":"nature-genetics","isVorOnly":false,"title":"Nature Genetics"},"publishedOn":"2025-09-05 04:00:00","publishedOnDateReadable":"September 5th, 2025"},"versionCreatedAt":"2024-07-03 06:26:10","video":"","vorDoi":"10.1038/s41588-025-02329-5","vorDoiUrl":"https://doi.org/10.1038/s41588-025-02329-5","workflowStages":[]},"version":"v1","identity":"rs-4631710","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4631710","identity":"rs-4631710","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}