key: cord-0309872-xhkox4pi authors: Han, Lei; Wei, Xiaoyu; Liu, Chuanyu; Volpe, Giacomo; Zhuang, Zhenkun; Zou, Xuanxuan; Wang, Zhifeng; Pan, Taotao; Yuan, Yue; Zhang, Xiao; Fan, Peng; Guo, Pengcheng; Lai, Yiwei; Lei, Ying; Liu, Xingyuan; Yu, Feng; Shangguan, Shuncheng; Lai, Guangyao; Deng, Qiuting; Liu, Ya; Wu, Liang; Shi, Quan; Yu, Hao; Huang, Yunting; Cheng, Mengnan; Xu, Jiangshan; Liu, Yang; Wang, Mingyue; Wang, Chunqing; Zhang, Yuanhang; Xie, Duo; Yang, Yunzhi; Yu, Yeya; Zheng, Huiwen; Wei, Yanrong; Huang, Fubaoqian; Lei, Junjie; Huang, Waidong; Zhu, Zhiyong; Lu, Haorong; Wang, Bo; Wei, Xiaofeng; Chen, Fengzhen; Yang, Tao; Du, Wensi; Chen, Jing; Xu, Shibo; An, Juan; Ward, Carl; Wang, Zongren; Pei, Zhong; Wong, Chi-Wai; Liu, Xiaolei; Zhang, Huafeng; Liu, Mingyuan; Qin, Baoming; Schambach, Axel; Isern, Joan; Feng, Liqiang; Liu, Yan; Guo, Xiangyu; Liu, Zhen; Sun, Qiang; Maxwell, Patrick H.; Barker, Nick; Muñoz-Cánoves, Pura; Gu, Ying; Mulder, Jan; Uhlen, Mathias; Tan, Tao; Liu, Shiping; Yang, Huanming; Wang, Jian; Hou, Yong; Xu, Xun; Esteban, Miguel A.; Liu, Longqi title: Cell transcriptomic atlas of the non-human primate Macaca fascicularis date: 2021-12-13 journal: bioRxiv DOI: 10.1101/2021.12.13.472311 sha: b2266330061548bc93f8c833935c4943674906b8 doc_id: 309872 cord_uid: xhkox4pi Studying tissue composition and function in non-human primates (NHP) is crucial to understand the nature of our own species. Here, we present a large-scale single-cell and single-nucleus transcriptomic atlas encompassing over one million cells from 43 tissues from the adult NHP Macaca fascicularis. This dataset provides a vast, carefully annotated, resource to study a species phylogenetically close to humans. As proof of principle, we have reconstructed the cell-cell interaction networks driving Wnt signalling across the body, mapped the distribution of receptors and co-receptors for viruses causing human infectious diseases and intersected our data with human genetic disease orthologous coordinates to identify both expected and unexpected associations. Our Macaca fascicularis cell atlas constitutes an essential reference for future single-cell studies in human and NHP. Guangzhou 510000, China. 48 9 muscle nuclei displayed subtype-specific and tissue-specific gene expression signatures 224 and gene ontology (GO) terms (Fig. 2e, f and Extended Data Fig 16a-c) . We also 225 noticed substantial myonuclei heterogeneity within the same subtype and tissue (Fig. 226 2f) . 227 Next, to explore the heterogeneity between different types of adipocytes, we 228 grouped and re-clustered cells from subcutaneous and visceral adipose tissues, resulting 229 in 10 major clusters (Extended Data Fig. 17a) . We observed a marked distinction 230 between mature adipocytes and adipocyte progenitors, as reflected by the differential 231 expression of ADIPOQ and CD34 (Extended Data Fig. 17b) . Visceral mature 232 adipocytes and adipocyte progenitors displayed enriched expression of ITLN1, in 233 agreement with visceral adipocytes having mesothelial origin 23 , and also high 234 mitochondrial activity exemplified by high expression of ND4, ATP6 and COX3 24,25 235 (Extended Data Fig. 17c, d) . In contrast, subcutaneous mature adipocytes and 236 adipocyte progenitors were enriched in FOS. Likewise, SLC11A1 and SPOCK3 marked 237 mature subcutaneous and visceral adipocytes, respectively. Adipocyte progenitors 238 contained two populations for visceral tissue (WT1 + and CFD high ), three for 239 subcutaneous tissue (ESR1 + , CXCL14 + APOD + and DPP4 + ) and one shared between 240 both tissues (NOX4 + ) (Extended Data Fig. 17a, c and d) . Within the subcutaneous 241 CXCL14 + APOD + progenitor cluster, we observed a population of CFD high cells that also 242 co-expressed DPP4, a marker of highly proliferative adipocyte progenitors in both 243 mouse and human 26 . However, we did not detect significant proliferation in any of the 244 monkey adipocyte progenitor populations based on the expression of the pan-cycling 245 marker MKI67 27 (Extended Data Fig. 17c) . NOX4 + is an NAPDH oxidase that acts as 246 a switch from insulin-induced proliferation to adipocyte differentiation, suggesting that 247 the shared cluster is a converging route for both adipose tissues towards adipocytic 248 maturation 28 . 249 Finally, we grouped and re-clustered all tissues that contain mesothelial cells, a type 250 of specialized epithelial cells. Mesothelial cells from bladder, ovary and fallopian tube 251 19 version of a large single-cell transcriptomic atlas for a NHP widely used in research 503 studies, Macaca fascicularis, and an expandable and interactive database 504 (https://db.cngb.org/nhpca/) to facilitate its exploration. The current version of our atlas 505 provides a comprehensive and integrated overview of gene expression in 106 cell types 506 extracted from 43 tissue types. Specialized tissues such as skin, thymus, testis and some 507 parts of the gastrointestinal tract, as well as increased cell numbers for some of the 508 already profiled ones, will be added in future releases. Cell type identification relied on 509 previously reported markers and gene expression profiles. Therefore, although we 510 identified most (if not all) known cell types in these tissues, our current annotations are 511 likely to benefit from deeper sub-clustering and further revision. 512 We provide a detailed description of individual tissue single-cell composition and 513 a comparison of common cell types across all sequenced tissues. This information will 514 be particularly valuable for understanding tissues that have either not been profiled at 515 all at the single-cell level in human (e.g., diaphragm, tongue and salivary gland) or lack 516 enough cell numbers (e.g., liver, gallbladder and substantia nigra), and for prediction 517 of human disease susceptibilities. Regarding the latter, we have identified an 518 unexpected link between ADHD and muscle function. ADHD is a polygenic and 519 multifactorial disorder associated with hyperactivity and motor coordination 520 abnormalities that are thought to have a neurological origin 76 . Our data support the 521 possibility that skeletal muscle rather than the nervous system may be a direct driver of 522 ADHD pathogenesis 77 . Similarly, as part of the analysis for virus receptors and co-523 receptors, we provide a comprehensive map of ACE2 + /TMPRSS2 + double positive cells 524 throughout the monkey body that may be useful to understand COVID-19 pathogenesis 525 in human 59,61 . In particular, the link between IL6, STAT transcription factors and ACE2 526 expression could explain the reported positive effects of tocilizumab, a humanized 527 monoclonal antibody against IL6R for the treatment of patients with severe COVID-528 19 78 . On the other hand, our study shows significant interspecies differences in cell 529 type-specific gene expression with potentially important functional consequences. For 530 20 example, the distribution of ACE2 and TMPRSS2 across different cell types is not 531 identical between monkey and human and this could influence the disease course. 532 Moreover, in the context of the survey of Wnt pathway components we have identified 533 LGR5 + renal cells with progenitor characteristics that are seemingly absent in human 534 and mouse based on analysis of reported datasets. This is relevant because the kidney 535 has limited regenerative capacity in mammals 79 . During embryonic development 536 LGR5 + cells located at the junction between the ureteric bud (source of the collecting 537 tubule and connecting tubule) and the metanephric blastema are responsible for 538 nephrogenesis, but they quickly disappear after birth 45 . Their persistence in adult 539 monkey kidney suggests a higher regenerative capacity compared to other species, 540 which if true raises the hope of activating a similar mechanism in human 80 . Similarly, 541 LGR5 + cells in the neocortex correspond mainly to OPC in monkey and to 542 oligodendrocytes and to a lesser extent OPC in human, whereas in mouse inhibitory 543 neurons are more highly enriched. This finding is consistent with the knowledge that 544 Wnt activity regulates OPC and oligodendrocyte function and differentiation 81 but 545 suggest interspecies differences in the mode of action. Likewise, the expression of 546 LGR5 in skeletal slow-twitch myofibers, and LGR6 in the pituitary gland and heart, is 547 intriguing. During development, Wnt activity regulates skeletal myogenesis and 548 myofiber typing 82 , cardiomyocyte proliferation 83 and pituitary gland growth 84 , but little 549 is known about the adult. The functional implications of these and other related findings 550 and the extent to which the patterns differ between monkey and other mammalian 551 species will require further study. Finally, interspecies comparison of single-cell gene 552 expression in neocortex highlights the problems associated with modelling neurological 553 diseases in rodents and suggests that a cautious approach should also be taken when 554 studying NHP. Additional comparisons with other human and mouse single-cell/nuclei 555 datasets will provide a more comprehensive, body-wide picture of differences in 556 disease vulnerability among the three species. 557 21 In the future, with efforts from us and scientists worldwide, the NHPCA database 558 will be extended with additional single-cell datasets generated from disease modelling 559 studies, spontaneously developed diseases (e.g., diabetes or cardiomyopathy) and aging. 560 Adding other layers of single-cell -omics studies, in particular scATAC-seq and 561 spatially resolved transcriptomics 85 for all tissues presented here, will help characterize 562 cell states and the interactions between different cell types more accurately. Proof of 563 principle is the kidney scATAC-seq dataset included here. In addition, it will be 564 important to compare our Macaca fascicularis atlas with datasets from other non-565 endangered NHP species such as Macaca mulatta (rhesus monkey), Callithrix jacchus 566 (marmoset monkey) 86 and Microcebus murinus (mouse lemur) 10,13 . Altogether, this 567 information will be instrumental for understanding primate evolution and human 568 disease. LGR6 and MKI67 in abdominal wall, adrenal gland, aorta, bladder, bone 858 marrow, bronchus, carotid, cerebellum, colon, diaphragm and duodenum. LGR6 and MKI67 in epididymis, esophagus, fallopian tube, gallbladder, heart, 865 kidney, liver, lung, lymph node and ovary. LGR6 and MKI67 in stomach, subcutaneous adipose tissue, substantia nigra, 880 thyroid, tongue, tonsil, trachea, uterus, vagina and visceral adipose tissue. Tissues were kept in an ice box and homogenized by 25-50 strokes of the loose pestle 1130 (Pestle A) after which the mixture was filtered using a 100 µm cell strainer in to a 1.5 1131 ml tube (Eppendorf). The mixture was then transferred to a clean 1 ml dounce 1132 homogenizer to which 750 ul of buffer A containing 1% Igepal (Sigma, #CA630) was 1133 added and the tissue was further homogenized by 25 strokes of the tight pestle (Pestle 1134 42 centrifuged at 500 g for five minutes at 4°C to pellet nuclei. At this stage, the pellet was 1136 resuspended in 1 ml of buffer B containing 320 mM Sucrose, 10 mg/ml BSA, 3 mM 1137 CaCl2, 2 mM MgAc2, 0.1 mM EDTA, 10 mM Tris-HCl, 1 mM DTT, 1x Protease 1138 Inhibitor and 0.12 U/µl RNasein. This was followed by a centrifugation at 500 g for 1139 five minutes at 4°C to pellet nuclei. Nuclei were then resuspended with cell 1140 resuspension buffer at a concentration of 1,000 nuclei/μl for single-nucleus library 1141 preparation. Cells from lymph node, spleen, duodenum, stomach and colon were 1142 Clustering analysis of the complete cynomolgus monkey tissue dataset was performed 1208 using Scanpy (version 1.6.0) 93 in a Python environment (version 3.6). Parameters used 1209 in each function were manually curated to portray the optimal clustering of cells. In the 1210 preprocessing, cells or nuclei were filtered based on the criteria of expressing a 1211 minimum of 500 genes and genes expressed by at least three cells or nuclei were kept 1212 for the following analysis. In addition, cells or nuclei with more than 10% 1213 mitochondrial gene counts were removed. Louvain method is then used to detect subgroups of cells. Distinguishing differential 1220 genes among clusters were ranked (Benjamini-Hochberg, Wilcoxon rank-sum test). 1221 Cell types were manually and iteratively assigned based on overlap of literature, curated 1222 and statistically ranked genes. Each tissue dataset was portrayed using the Seurat 1223 package (version 3.2.2) 94 in R environment (version 3.6). Data from different replicates 1224 were integrated following the standard integrated pipeline by default parameters for 1225 filtering, data normalization, dimensionality reduction, clustering and gene differential 1226 In the global clustering, we performed DEG analysis using the sc.pl.rank_genes_groups 1232 function in Scanpy (V1.6.0). In other studies, we used the FindMarker or 1233 FindAllMarker function in the Seurat R package (V3.2.2). Analysis of DEG comparing 1234 specific populations was performed by calculating the fold-change of the mean 1235 expression level of genes between the selected populations. DEG were defined as those 1236 with a fold-change > 2 and adjusted P < 0.01. GO enrichment analysis was performed 1237 using the CompareCluster function fun = "enrichGO", pvalueCutoff = 0.1, 1238 pAdjustMethod = "BH", OrgDb = org.Hs.eg.db,ontBP") of ChIPseeker R package 1239 (v.1.22.1) 95 . Only GO terms with adjusted P < 0.05 were retained. 1240 1241 For tissue inter-species analysis, in order to get more accurate comparisons, we 1243 specifically chose three tissues with snRNA-seq data, namely kidney, neocortex and 1244 heart, and processed the raw sequencing data using our pipeline described below in the 1245 'Sc/snRNA-seq data processing' section. Kidney 43,44 , neocortex 46 and heart 49,50 data 1246 46 were downloaded from NCBI Gene expression omnibus (human kidney: GSE121862, mouse kidney: GSE119531, human neocortex: GSE97942, human heart: ERP123138, 1248 mouse heart: E-MTAB-7869). For each tissue we preprocessed the UMI matrix of the 1249 three species following three steps: 1. only orthologs genes among three species were 1250 kept. 2. only genes expressed in at least one cell in one species were kept. 3. the gene 1251 names of the human and mouse UMI matrix were converted into orthologs in Macaca 1252 fascicularis. After preprocessing, the UMI matrices of the three species were integrated 1253 together and the clustering was performed following the standard integrated pipeline 1254 using Seurat (V3.2.2) with the addition of one additional criterion for which only cells 1255 expressing more than 500 genes were kept. Also, we downsampled the cells of human 1256 and macaque neocortex to 10,000 to get a better clustering result. The Seurat clusters 1257 were then annotated into different cell types using cell type-specific markers as 1258 described above. In addition, for the comparison presented in Extended Data Figure 35 1259 we retrieved the publicly available single-cell data for gallbladder, liver and lung from 1260 GEO GSE134355 3 , GEO GSE108098 6 and GSE124395 96 , respectively. Data from the 1261 three species were integrated, clustered and annotated in the same way as described. Nuclei with TSS enrichment score lower than five and fragment number lower than 1313 1,000 were removed. Then, we calculated the doublet score with addDoubletScores 1314 function in ArchR package and filtered doublets by filterDoublets function with 1315 parameter filterRatio = 2. ScATAC-seq clustering analysis was performed using ArchR 1316 software by first identifying a robust set of peak regions followed by iterative LSI 1317 (latent semantic indexing) clustering. Briefly, we created 500 bp windows tiled across 1318 the genome and determined whether each cell was accessible within each window. Next, 1319 we performed an LSI dimensionality reduction on these windows with addIterativeLSI 1320 function in ArchR packages. We then performed Seurat clustering (FindClusters) on 1321 the LSI dimensions at resolutions of 0.8. Anchors between scATAC-seq and 1322 sc/snRNA-seq datasets were identified and used to transfer cell type labels identified 1323 from the sc/snRNA-seq data. We embedded the data by the TransferData function of 1324 The Human Cell Atlas: from vision to reality. 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 ADIPOQ COL5A3 EMCN FLT1 MYH11 PDGFRB CYP4B1 AGER SFTPB SFTPC SLC12A1 EGF SLC1A2 GFAP MS4A1 CD74 FCRL1 ARHGAP15 UPK2 UPK3A HBB RAG2 AIF1 P2RY6 ESM1 CFD TNNI3 MYBPC3 GABRG1 SLC1A3 CX3CR1 P2RY12 PLP1 MOBP CHGA TH SCGB3A1 SCGB1A1 PKHD1 SLC8A1 POMC GH1 ITLN1 PKHD1L1 SLC12A3 FXYD2 EMCN PECAM1 FLT1 PECAM1 OLIG1 CLDN5 KRT5 TP63 AGTR2 MYH11 ADAM7 OVCH2 HBM GATA1 NEB MYBPC1 KRT13 KRT4 S100A9 DCN NEUROD6 CAMK2A SCGB1A1 PAX8 CYP11B1 CYP17A1 FOXJ1 DNAH5 KLK3 DCN FSHB GNRHR IL22 IGKC CFD DCN SCGB3A1 SLC26A3 SLC13A1 SCGB3A1 CYP11B2 CYP21A2 GABRA6 CBLN1 CRHBP FSHR COLEC10 HGF GAD1 GAD2 FLT1 PECAM1 ATP6V0D2 FOXI1 PRL POU1F1 STAB2 FCN2 CD163 SIGLEC1 TP63 KRT5 PNMA6E ACTG2 SLC26A4 ATP6V0D2 MLANA TYRP1 ITLN1 PKHD1L1 GAD1 SORCS3 LIPF TFF2 RLBP1 WIF1 NEFM NEFH OLIG2 PDGFRA MOG OPALIN COL1A2 LAMA2 CPA2 CPB1 GCG IAPP INS PNLIP LAMA2 DCN RHO GNAT1 TPH1 SLC5A7 EMCN CLDN9 JCHAIN MZB1 NPHS2 PTPRO FXYD4 HSD11B2 SEMG2 PATE4 SLC34A1 LRP2 DNAH5 FOXJ1 FCN3 VEGFC MRC1 TREM1 RBPMS SLC17A6 RGR RPE65 CYP2B6 CPS1 ALB PLG SMR3B PRR27 KRT14 KRT17 KLK1 ATP6V0A4 BPIFB2 KRT81 MYH11 ACTG2 GH1 GHRHR SEMG2 DEFB126 CRISP1 RNASE9 DCN FBN1 CD3D CD3G IL7R CTLA4 SPRR2G KRT16 EPB42 WT1 TG TPO TSHB NELL1 NEB MYH2 MYH7 TNNT1 MYH2 MYBPC2 PRR27 LPO SLC6A2 PGR KLK3 MYH11 ITLN1 PKHD1L1 CYP11B1 HSD3B2 CACNB2 KHDRBS2 AXDND1 CYP11A STAR CYP17A1 GRAMD1B CALN1 CAPN13 KRT19 DHRS2 SPINK1 SNX31 TBX1 TMEM132D PKHD1L1 MMRN1 NTN1 LRRIQ1 DNAH5 DNAH9 VMO1 DKK2 PDE3A PTGIS NEK10 BMX EXT1 MEOX2 ZNF385D GUCY1A3 PLCB4 HMCN1 ABI3BP EFNA5 ENPP3 NEB TTN MYH2 MYBPC1 TRDN ACTA1 TNNC2 CMYA5 MYPN DMD HBB FTH1 IGKC UBB ACTB RPS20 RPL19 RPS7 TPT1 SPARC ADAM7 OVCH2 DEFB128 ADGRG2 GPX5 ADAM28 CST11 WFDC9 DEFB118 LCN9 KRT13 KRT4 S100A9 CSTB ECM1 SPINK5 CRNN KRT78 TGM3 SPINK7 SCGB3A1 AGBL4 UNC13C CTNND2 ANXA4 SLC13A1 CAPN12 SLC4A4 PIK3C2G PLD1 TINAG MIOX CUBN SLC12A1 PKHD1 LRP2 RYR2 EGF BICC1 STAB2 ALB FCN2 STAB1 CPS1 EDA MRC1 BMPER OIT3 APOA2 AFF3 PRICKLE2 RTKN2 SFTPC HPGD NCKAP5 VEGFC NXN SLCO2A1 EPAS1 GRB14 APOA1 IGFBP2 CRHBP RBFOX1 APOD FHOD3 GUCY1A2 DIAPH3 GPIHBP1 MYL1 PDLIM3 TRPM1 RPE65 GRID2 TRPM3 NKAIN2 SEMA3A ABCG2 VWF COL1A1 GH1 PRL POMC CGA CHGB GHRHR DACH2 KCNIP4 ROBO2 BPIFA2 SMR3B PIP MUC7 CA6 STATH KLK1 PRR27 CFTR LPO PGC FABP5 LIPF CD320 RGS5 TMSB10 IGFBP3 DCN SULF1 GSN LHFPL6 CDH13 PLCXD3 TG TPO PAX8 SLA NTM IYD SLC26A7 KIAA1456 HS6ST3 KCNQ5 COX1 ND3 CNN3 ND2 CYTB ND1 PTPRC FCRL1 S100A8 ARHGAP15 BANK1 STK17B SKAP1 PRUNE2 MYH11 PALLD PGR ADAMTS19 COL3A1 ENPP2 RORB LPCAT2 EBF3 LYPD6 FMO2 PLXNA4 DSCAM ENPP1 EBF1 ITLN1 FABP4 SCD DGAT2 CFD PPARG GPAM LPL LGR5 LGR6 MKI67 LGR5 LGR6 MKI67 LGR5 LGR6 MKI67 LGR5 (human kidney) Human kidney (9,726 nuclei) Monkey kidney (62,873 nuclei) Mouse kidney (2,052 nuclei) LGR5 (monkey kidney) LGR5 (mouse kidney) LGR5 (human neocortex) Human neocortex (10,000 nuclei) Monkey neocortex (10,000 nuclei) Mouse neocortex (9,206 nuclei) LGR5 (monkey neocortex) LGR5 (mouse neocortex) LGR6 (monkey heart) LGR6 (mouse heart) UMAP 1 LGR5 LGR6 RSPO1 RSPO2 RSPO3 RSPO4 WNT1 WNT2 WNT2B WNT3 WNT3A WNT4 WNT5A WNT5B WNT6 WNT7A WNT7B WNT8A WNT8B WNT9A WNT9B WNT10A WNT10B WNT11 WNT16 FZD2 FZD3 FZD4 FZD5 FZD6 FZD7 FZD8 FZD9 FZD10 LRP5 LRP6 RYK MUSK PTK7 ROR1 ROR2 TCF7 LEF1 TCF7L1 TCF7L2 WNT1 WNT2 WNT2B WNT3 WNT3A WNT4 WNT5A WNT5B WNT6 WNT7A WNT7B WNT8A WNT8B WNT9A WNT9B WNT10A WNT10B WNT11 WNT16 FZD2 FZD3 FZD4 FZD5 FZD6 FZD7 FZD8 FZD9 FZD10 LRP5 LRP6 RYK MUSK PTK7 ROR1 ROR2 RYK TCF7 LEF1 TCF7L1 TCF7L2 LGR5 LGR6 RSPO1 RSPO2 RSPO3 RSPO4 WNT1 WNT2 WNT2B WNT3 WNT3A WNT4 WNT5A WNT5B WNT6 WNT7A WNT7B WNT8A WNT8B WNT9A WNT9B WNT10A WNT10B WNT11 WNT16 FZD2 FZD3 FZD4 FZD5 FZD6 FZD7 FZD8 FZD9 FZD10 LRP5 LRP6 MUSK PTK7 ROR1 ROR2 TCF7 LEF1 TCF7L1 LGR5 LGR6 RSPO1 RSPO2 RSPO3 RSPO4 WNT1 WNT2 WNT2B WNT3 WNT3A WNT4 WNT5A WNT5B WNT6 WNT7A WNT7B WNT8A WNT8B WNT9A WNT9B WNT10A WNT10B WNT11 WNT16 FZD2 FZD3 FZD4 FZD5 FZD6 FZD7 FZD8 FZD9 FZD10 LRP5 LRP6 RYK MUSK PTK7 ROR1 ROR2 TCF7 LEF1 TCF7L1 TCF7L2 Ligand RSPO1-4 : Receptor LGR5,6 Ligand WNT factors : Receptor FZD etc Liver 3 0 2 0 10 0 LGR5 LGR6 RSPO1 RSPO2 RSPO3 RSPO4 WNT1 WNT2 WNT2B WNT3 WNT3A WNT4 WNT5A WNT5B WNT6 WNT7A WNT7B WNT8A WNT8B WNT9A WNT9B WNT10A WNT10B WNT11 WNT16 FZD2 FZD3 FZD4 FZD5 FZD6 FZD7 FZD8 FZD9 FZD10 LRP5 LRP6 RYK MUSK PTK7 ROR1 ROR2 TCF7 LEF1 TCF7L1 LGR5 LGR6 RSPO1 RSPO2 RSPO3 RSPO4 WNT1 WNT2 WNT2B WNT3 WNT3A WNT4 WNT5A WNT5B WNT6 WNT7A WNT7B WNT8A WNT8B WNT9A WNT9B WNT10A WNT10B WNT11 WNT16 FZD2 FZD3 FZD4 FZD5 FZD6 FZD7 FZD8 FZD9 FZD10 LRP5 LRP6 RYK MUSK PTK7 ROR1 ROR2 TCF7 LEF1 TCF7L1 LGR5 LGR6 RSPO1 RSPO2 RSPO3 RSPO4 WNT1 WNT2 WNT2B WNT3 WNT3A WNT4 WNT5A WNT5B WNT6 WNT7A WNT7B WNT8A WNT8B WNT9A WNT9B WNT10A WNT10B WNT11 WNT16 FZD2 FZD3 FZD4 FZD5 FZD6 FZD7 FZD9 FZD8 FZD10 LRP5 LRP6 RYK MUSK PTK7 ROR1 ROR2 TCF7 LEF1 TCF7L1 LGR5 LGR6 RSPO1 RSPO2 RSPO3 RSPO4 WNT1 WNT2 WNT2B WNT3 WNT3A WNT4 WNT5A WNT5B WNT6 WNT7A WNT7B WNT8A WNT8B WNT9A WNT9B WNT10A WNT10B WNT11 WNT16 FZD2 FZD3 FZD4 FZD5 FZD6 FZD7 FZD8 FZD9 FZD10 LRP5 LRP6 RYK MUSK PTK7 ROR1 ROR2 TCF7 LEF1 TCF7L1 LGR5 LGR6 RSPO1 RSPO2 RSPO3 RSPO4 WNT1 WNT2 WNT2B WNT3 WNT3A WNT4 WNT5A WNT5B WNT6 WNT7A WNT7B WNT8A WNT8B WNT9A WNT9B WNT10A WNT10B WNT11 WNT16 FZD2 FZD3 FZD4 FZD5 FZD6 FZD7 FZD8 FZD9 FZD10 LRP5 LRP6 RYK MUSK PTK7 ROR1 ROR2 TCF7 LEF1 TCF7L1 Systematic comparison of single-cell and single-nucleus RNA-1425 sequencing methods Structural cells are key regulators of organ-specific 1428 immune responses Single-Cell Transcriptome Atlas of Murine Endothelial Cells Cross-Species Single-Cell Analysis Reveals Divergence of 1433 the Primate Microglia Program Single-nucleus RNA-seq identifies transcriptional 1436 heterogeneity in multinucleated skeletal myofibers Human tongue neuroanatomy: Nerve supply and motor 1439 endplates Visceral and subcutaneous fat have different origins and 1441 evidence supports a mesothelial source Single-Cell Transcriptomic Atlas of Primate Ovarian Aging Single-cell-resolution transcriptome map of human, 1580 chimpanzee, bonobo, and macaque brains Motor regulation problems and pain in adults diagnosed with 1583 ADHD Genome-wide association study of motor coordination 1585 problems in ADHD identifies genes for brain and muscle function Tocilizumab for induction and maintenance of remission 1588 in giant cell arteritis: a phase 2, randomised, double-blind, placebo-controlled 1589 trial Concise Reviews: Stem 1591 Cells and Kidney Regeneration: An Update Recreating, expanding and using nephron 1594 progenitor populations Wnt-Dependent Oligodendroglial-Endothelial Interactions 1597 Regulate White Matter Vascularization and Attenuate Injury Wnt Signaling in Skeletal Muscle Development and 1600 Regeneration Hippo pathway inhibits Wnt signaling to restrain 1603 cardiomyocyte proliferation and heart size Canonical 1606 WNT Signaling Regulates the Pituitary Organizer and Pituitary Gland 1607 Formation Large field of view-spatially resolved transcriptomics at 1609 nanoscale resolution. bioRxiv The marmoset monkey as a model for visual 1612 neuroscience Single-nucleus and single-cell transcriptomes compared in 1615 matched cortical cell types Cell-Type Specificity of Genomic Imprinting in Cerebral 1618 Cortex STAR: ultrafast universal RNA-seq aligner Sambamba: fast 1623 processing of NGS alignment formats A single-nuclei RNA sequencing study of Mendelian 1626 and sporadic AD in the human brain Doublet 1629 Detection in Single-Cell RNA Sequencing Data Using Artificial Nearest 1630 Neighbors SCANPY: large-scale single-cell gene 1632 expression data analysis Comprehensive Integration of Single-Cell Data ChIPseeker: an R/Bioconductor package for 1637 ChIP peak annotation, comparison and visualization A human liver cell atlas reveals heterogeneity and epithelial 1640 progenitors The dynamics and regulators of cell fate decisions are 1642 revealed by pseudotemporal ordering of single cells CellPhoneDB: inferring cell-cell communication from combined expression of 1646 multi-subunit ligand-receptor complexes ArchR is a scalable software package for integrative single-1649 cell chromatin accessibility analysis SeqKit: A Cross-Platform and Ultrafast 1652 Toolkit for FASTA/Q File Manipulation