FUSION : A web-based application for in-depth exploration of multi-omics data with brightfield histology.

Spatial -OMICS technologies facilitate the interrogation of molecular profiles in the context of the underlying histopathology and tissue microenvironment. Paired analysis of histopathology and molecular data can provide pathologists with otherwise unobtainable insights into biological mechanisms. To connect the disparate molecular and histopathologic features into a single workspace, we developed FUSION ( F unctional U nit S tate I dentificati ON in WSIs [Whole Slide Images]), a web-based tool that provides users with a broad array of visualization and analytical tools including deep learning-based algorithms for in-depth interrogation of spatial -OMICS datasets and their associated high-resolution histology images. FUSION enables end-to-end analysis of functional tissue units (FTUs), automatically aggregating underlying molecular data to provide a histopathology-based medium for analyzing healthy and altered cell states and driving new discoveries using "pathomic" features. We demonstrate FUSION using 10x Visium spatial transcriptomics (ST) data from both formalin-fixed paraffin embedded (FFPE) and frozen prepared datasets consisting of healthy and diseased tissue. Through several use-cases, we demonstrate how users can identify spatial linkages between quantitative pathomics, qualitative image characteristics, and spatial --omics.

Protein N-glycans in Healthy and Sclerotic Glomeruli in Diabetic Kidney Disease.

BACKGROUND: Diabetes is expected to directly impact renal glycosylation, yet to date, there has not been a comprehensive evaluation of alterations in N-glycan composition in the glomeruli of patients with diabetic kidney disease (DKD). METHODS: We used untargeted mass spectrometry imaging to identify N-glycan structures in healthy and sclerotic glomeruli in FFPE sections from needle biopsies of five patients with DKD and three healthy kidney samples. Regional proteomics was performed on glomeruli from additional biopsies from the same patients to compare the abundances of enzymes involved in glycosylation. Secondary analysis of single nuclei transcriptomics (snRNAseq) data was used to inform on transcript levels of glycosylation machinery in different cell types and states. RESULTS: We detected 120 N-glycans, and among them identified twelve of these protein post-translated modifications that were significantly increased in glomeruli. All glomeruli-specific N-glycans contained an N-acetyllactosamine (LacNAc) epitope. Five N-glycan structures were highly discriminant between sclerotic and healthy glomeruli. Sclerotic glomeruli had an additional set of glycans lacking fucose linked to their core, and they did not show tetra-antennary structures that are common in healthy glomeruli. Orthogonal omics analyses revealed lower protein abundance and lower gene expression involved in synthesizing fucosylated and branched N-glycans in sclerotic podocytes. In snRNAseq and regional proteomics analyses, we observed that genes and/or proteins involved in sialylation and LacNAc synthesis were also downregulated in DKD glomeruli, but this alteration remained undetectable by our spatial N-glycomics assay. CONCLUSIONS: Integrative spatial glycomics, proteomics, and transcriptomics revealed protein N-glycosylation characteristic of sclerotic glomeruli in DKD.

ComPRePS: An Automated Cloud-based Image Analysis tool to democratize AI in Digital Pathology.

Artificial intelligence (AI) has extensive applications in a wide range of disciplines including healthcare and clinical practice. Advances in high-resolution whole-slide brightfield microscopy allow for the digitization of histologically stained tissue sections, producing gigapixel-scale whole-slide images (WSI). The significant improvement in computing and revolution of deep neural network (DNN)-based AI technologies over the last decade allow us to integrate massively parallelized computational power, cutting-edge AI algorithms, and big data storage, management, and processing. Applied to WSIs, AI has created opportunities for improved disease diagnostics and prognostics with the ultimate goal of enhancing precision medicine and resulting patient care. The National Institutes of Health (NIH) has recognized the importance of developing standardized principles for data management and discovery for the advancement of science and proposed the Findable, Accessible, Interoperable, Reusable, (FAIR) Data Principles1 with the goal of building a modernized biomedical data resource ecosystem to establish collaborative research communities. In line with this mission and to democratize AI-based image analysis in digital pathology, we propose ComPRePS: an end-to-end automated Computational Renal Pathology Suite which combines massive scalability, on-demand cloud computing, and an easy-to-use web-based user interface for data upload, storage, management, slide-level visualization, and domain expert interaction. Moreover, our platform is equipped with both in-house and collaborator developed sophisticated AI algorithms in the back-end server for image analysis to identify clinically relevant micro-anatomic functional tissue units (FTU) and to extract image features.

ComPRePS: An Automated Cloud-based Image Analysis tool to democratize AI in Digital Pathology.

Artificial intelligence (AI) has extensive applications in a wide range of disciplines including healthcare and clinical practice. Advances in high-resolution whole-slide brightfield microscopy allow for the digitization of histologically stained tissue sections, producing gigapixel-scale whole-slide images (WSI). The significant improvement in computing and revolution of deep neural network (DNN)-based AI technologies over the last decade allow us to integrate massively parallelized computational power, cutting-edge AI algorithms, and big data storage, management, and processing. Applied to WSIs, AI has created opportunities for improved disease diagnostics and prognostics with the ultimate goal of enhancing precision medicine and resulting patient care. The National Institutes of Health (NIH) has recognized the importance of developing standardized principles for data management and discovery for the advancement of science and proposed the Findable, Accessible, Interoperable, Reusable, (FAIR) Data Principles1 with the goal of building a modernized biomedical data resource ecosystem to establish collaborative research communities. In line with this mission and to democratize AI-based image analysis in digital pathology, we propose ComPRePS: an end-to-end automated Computational Renal Pathology Suite which combines massive scalability, on-demand cloud computing, and an easy-to-use web-based user interface for data upload, storage, management, slide-level visualization, and domain expert interaction. Moreover, our platform is equipped with both in-house and collaborator developed sophisticated AI algorithms in the back-end server for image analysis to identify clinically relevant micro-anatomic functional tissue units (FTU) and to extract image features.

The chromatin landscape of healthy and injured cell types in the human kidney.

There is a need to define regions of gene activation or repression that control human kidney cells in states of health, injury, and repair to understand the molecular pathogenesis of kidney disease and design therapeutic strategies. Comprehensive integration of gene expression with epigenetic features that define regulatory elements remains a significant challenge. We measure dual single nucleus RNA expression and chromatin accessibility, DNA methylation, and H3K27ac, H3K4me1, H3K4me3, and H3K27me3 histone modifications to decipher the chromatin landscape and gene regulation of the kidney in reference and adaptive injury states. We establish a spatially-anchored epigenomic atlas to define the kidney's active, silent, and regulatory accessible chromatin regions across the genome. Using this atlas, we note distinct control of adaptive injury in different epithelial cell types. A proximal tubule cell transcription factor network of ELF3, KLF6, and KLF10 regulates the transition between health and injury, while in thick ascending limb cells this transition is regulated by NR2F1. Further, combined perturbation of ELF3, KLF6, and KLF10 distinguishes two adaptive proximal tubular cell subtypes, one of which manifested a repair trajectory after knockout. This atlas will serve as a foundation to facilitate targeted cell-specific therapeutics by reprogramming gene regulatory networks.

Multi-omics data integration shines a light on the renal medulla.

The renal medulla maintains salt and water balance and is prone to dysregulation because of high oxygen demand. Challenges in obtaining high-quality tissue have limited characterization of molecular programs regulating the medulla. Haug et al. leveraged gene expression, chromatin accessibility, long-range chromosomal interactions, and spatial transcriptomics to build a reference set of medullary tissue marker genes to define the medullary role in kidney function, exemplifying the strength and utility of multi-omic data integration.

Unbiased kidney-centric molecular categorization of chronic kidney disease as a step towards precision medicine.

Current classification of chronic kidney disease (CKD) into stages using indirect systemic measures (estimated glomerular filtration rate (eGFR) and albuminuria) is agnostic to the heterogeneity of underlying molecular processes in the kidney thereby limiting precision medicine approaches. To generate a novel CKD categorization that directly reflects within kidney disease drivers we analyzed publicly available transcriptomic data from kidney biopsy tissue. A Self-Organizing Maps unsupervised artificial neural network machine-learning algorithm was used to stratify a total of 369 patients with CKD and 46 living kidney donors as healthy controls. Unbiased stratification of the discovery cohort resulted in identification of four novel molecular categories of disease termed CKD-Blue, CKD-Gold, CKD-Olive, CKD-Plum that were replicated in independent CKD and diabetic kidney disease datasets and can be further tested on any external data at kidneyclass.org. Each molecular category spanned across CKD stages and histopathological diagnoses and represented transcriptional activation of distinct biological pathways. Disease progression rates were highly significantly different between the molecular categories. CKD-Gold displayed rapid progression, with significant eGFR-adjusted Cox regression hazard ratio of 5.6 [1.01-31.3] for kidney failure and hazard ratio of 4.7 [1.3-16.5] for composite of kidney failure or a 40% or more eGFR decline. Urine proteomics revealed distinct patterns between the molecular categories, and a 25-protein signature was identified to distinguish CKD-Gold from other molecular categories. Thus, patient stratification based on kidney tissue omics offers a gateway to non-invasive biomarker-driven categorization and the potential for future clinical implementation, as a key step towards precision medicine in CKD.

Blood immunophenotyping identifies distinct kidney histopathology and outcomes in patients with lupus nephritis.

Lupus nephritis (LN) is a frequent manifestation of systemic lupus erythematosus, and fewer than half of patients achieve complete renal response with standard immunosuppressants. Identifying non-invasive, blood-based pathologic immune alterations associated with renal injury could aid therapeutic decisions. Here, we used mass cytometry immunophenotyping of peripheral blood mononuclear cells in 145 patients with biopsy-proven LN and 40 healthy controls to evaluate the heterogeneity of immune activation in patients with LN and to identify correlates of renal parameters and treatment response. Unbiased analysis identified 3 immunologically distinct groups of patients with LN that were associated with different patterns of histopathology, renal cell infiltrates, urine proteomic profiles, and treatment response at one year. Patients with enriched circulating granzyme B+ T cells at baseline showed more severe disease and increased numbers of activated CD8 T cells in the kidney, yet they had the highest likelihood of treatment response. A second group characterized primarily by a high type I interferon signature had a lower likelihood of response to therapy, while a third group appeared immunologically inactive by immunophenotyping at enrollment but with chronic renal injuries. Main immune profiles could be distilled down to 5 simple cytometric parameters that recapitulate several of the associations, highlighting the potential for blood immune profiling to translate to clinically useful non-invasive metrics to assess immune-mediated disease in LN.

Genetic Analysis of Obesity-Induced Diabetic Nephropathy in BTBR Mice.

Diabetic nephropathy (DN) is the leading cause of end-stage renal disease in the U.S. and has a significant impact on human suffering. Leptin-deficient BTBR (BTBRob/ob) mice develop hallmark features of obesity-induced DN, whereas leptin-deficient C57BL/6J (B6ob/ob) mice do not. To identify genetic loci that underlie this strain difference, we constructed an F2 intercross between BTBRob/ob and B6ob/ob mice. We isolated kidneys from 460 F2 mice and histologically scored them for percent mesangial matrix and glomerular volume (∼50 glomeruli per mouse), yielding ∼45,000 distinct measures in total. The same histological measurements were made in kidneys from B6 and BTBR mice, either lean or obese (Lepob/ob), at 4 and 10 weeks of age, allowing us to assess the contribution of strain, age, and obesity to glomerular pathology. All F2 mice were genotyped for ∼5,000 single nucleotide polymorphisms (SNPs), ∼2,000 of which were polymorphic between B6 and BTBR, enabling us to identify a quantitative trait locus (QTL) on chromosome 7, with a peak at ∼30 Mbp, for percent mesangial matrix, glomerular volume, and mesangial volume. The podocyte-specific gene nephrin (Nphs1) is physically located at the QTL and contains high-impact SNPs in BTBR, including several missense variants within the extracellular immunoglobulin-like domains.

Scaffold protein SH3BP2 signalosome is pivotal for immune activation in nephrotic syndrome.

Despite clinical use of immunosuppressive agents, the immunopathogenesis of minimal change disease (MCD) and focal segmental glomerulosclerosis (FSGS) remains unclear. Src homology 3-binding protein 2 (SH3BP2), a scaffold protein, forms an immune signaling complex (signalosome) with 17 other proteins, including phospholipase Cγ2 (PLCγ2) and Rho-guanine nucleotide exchange factor VAV2 (VAV2). Bioinformatic analysis of human glomerular transcriptome (Nephrotic Syndrome Study Network cohort) revealed upregulated SH3BP2 in MCD and FSGS. The SH3BP2 signalosome score and downstream MyD88, TRIF, and NFATc1 were significantly upregulated in MCD and FSGS. Immune pathway activation scores for Toll-like receptors, cytokine-cytokine receptor, and NOD-like receptors were increased in FSGS. Lower SH3BP2 signalosome score was associated with MCD, higher estimated glomerular filtration rate, and remission. Further work using Sh3bp2KI/KI transgenic mice with a gain-in-function mutation showed ~6-fold and ~25-fold increases in albuminuria at 4 and 12 weeks, respectively. Decreased serum albumin and unchanged serum creatinine were observed at 12 weeks. Sh3bp2KI/KI kidney morphology appeared normal except for increased mesangial cellularity and patchy foot process fusion without electron-dense deposits. SH3BP2 co-immunoprecipitated with PLCγ2 and VAV2 in human podocytes, underscoring the importance of SH3BP2 in immune activation. SH3BP2 and its binding partners may determine the immune activation pathways resulting in podocyte injury leading to loss of the glomerular filtration barrier.

AAV-SPL 2.0, a Modified Adeno-Associated Virus Gene Therapy Agent for the Treatment of Sphingosine Phosphate Lyase Insufficiency Syndrome.

Sphingosine-1-phosphate lyase insufficiency syndrome (SPLIS) is an inborn error of metabolism caused by inactivating mutations in SGPL1, the gene encoding sphingosine-1-phosphate lyase (SPL), an essential enzyme needed to degrade sphingolipids. SPLIS features include glomerulosclerosis, adrenal insufficiency, neurological defects, ichthyosis, and immune deficiency. Currently, there is no cure for SPLIS, and severely affected patients often die in the first years of life. We reported that adeno-associated virus (AAV) 9-mediated SGPL1 gene therapy (AAV-SPL) given to newborn Sgpl1 knockout mice that model SPLIS and die in the first few weeks of life prolonged their survival to 4.5 months and prevented or delayed the onset of SPLIS phenotypes. In this study, we tested the efficacy of a modified AAV-SPL, which we call AAV-SPL 2.0, in which the original cytomegalovirus (CMV) promoter driving the transgene is replaced with the synthetic "CAG" promoter used in several clinically approved gene therapy agents. AAV-SPL 2.0 infection of human embryonic kidney (HEK) cells led to 30% higher SPL expression and enzyme activity compared to AAV-SPL. Newborn Sgpl1 knockout mice receiving AAV-SPL 2.0 survived ≥ 5 months and showed normal neurodevelopment, 85% of normal weight gain over the first four months, and delayed onset of proteinuria. Over time, treated mice developed nephrosis and glomerulosclerosis, which likely resulted in their demise. Our overall findings show that AAV-SPL 2.0 performs equal to or better than AAV-SPL. However, improved kidney targeting may be necessary to achieve maximally optimized gene therapy as a potentially lifesaving SPLIS treatment.

miRNA and mRNA Signatures in Human Acute Kidney Injury Tissue.

Acute kidney injury (AKI) is an important contributor to the development of chronic kidney disease (CKD). There is a need to understand molecular mediators that drive either recovery or progression to CKD. In particular, the role of miRNA and its regulatory role in AKI is poorly understood. We performed miRNA and mRNA sequencing on biobanked human kidney tissues obtained in the routine clinical care of patients with the diagnoses of AKI and minimal change disease (MCD), in addition to nephrectomized (Ref) tissue from individuals without known kidney disease. Transcriptomic analysis of mRNA revealed that Ref tissues exhibited a similar injury signature to AKI, not identified in MCD samples. The transcriptomic signature of human AKI was enriched with genes in pathways involved in cell adhesion and epithelial-to-mesenchymal transition (e.g., CDH6, ITGB6, CDKN1A ). miRNA DE analysis revealed upregulation of miRNA associated with immune cell recruitment and inflammation (e.g., miR-146a, miR-155, miR-142, miR-122). These miRNA (i.e., miR-122, miR-146) are also associated with downregulation of mRNA such as DDR2 and IGFBP6 , respectively. These findings suggest integrated interactions between miRNAs and target mRNAs in AKI-related processes such as inflammation, immune cell activation and epithelial-to-mesenchymal transition. These data contribute several novel findings when describing the epigenetic regulation of AKI by miRNA, and also underscores the importance of utilizing an appropriate reference control tissue to understand canonical pathway alterations in AKI.

Endogenous adenine mediates kidney injury in diabetic models and predicts diabetic kidney disease in patients.

Diabetic kidney disease (DKD) can lead to end-stage kidney disease (ESKD) and mortality; however, few mechanistic biomarkers are available for high-risk patients, especially those without macroalbuminuria. Urine from participants with diabetes from the Chronic Renal Insufficiency Cohort (CRIC) study, the Singapore Study of Macro-angiopathy and Micro-vascular Reactivity in Type 2 Diabetes (SMART2D), and the American Indian Study determined whether urine adenine/creatinine ratio (UAdCR) could be a mechanistic biomarker for ESKD. ESKD and mortality were associated with the highest UAdCR tertile in the CRIC study and SMART2D. ESKD was associated with the highest UAdCR tertile in patients without macroalbuminuria in the CRIC study, SMART2D, and the American Indian study. Empagliflozin lowered UAdCR in nonmacroalbuminuric participants. Spatial metabolomics localized adenine to kidney pathology, and single-cell transcriptomics identified ribonucleoprotein biogenesis as a top pathway in proximal tubules of patients without macroalbuminuria, implicating mTOR. Adenine stimulated matrix in tubular cells via mTOR and stimulated mTOR in mouse kidneys. A specific inhibitor of adenine production was found to reduce kidney hypertrophy and kidney injury in diabetic mice. We propose that endogenous adenine may be a causative factor in DKD.

Clustering-based spatial analysis (CluSA) framework through graph neural network for chronic kidney disease prediction using histopathology images.

Machine learning applied to digital pathology has been increasingly used to assess kidney function and diagnose the underlying cause of chronic kidney disease (CKD). We developed a novel computational framework, clustering-based spatial analysis (CluSA), that leverages unsupervised learning to learn spatial relationships between local visual patterns in kidney tissue. This framework minimizes the need for time-consuming and impractical expert annotations. 107,471 histopathology images obtained from 172 biopsy cores were used in the clustering and in the deep learning model. To incorporate spatial information over the clustered image patterns on the biopsy sample, we spatially encoded clustered patterns with colors and performed spatial analysis through graph neural network. A random forest classifier with various groups of features were used to predict CKD. For predicting eGFR at the biopsy, we achieved a sensitivity of 0.97, specificity of 0.90, and accuracy of 0.95. AUC was 0.96. For predicting eGFR changes in one-year, we achieved a sensitivity of 0.83, specificity of 0.85, and accuracy of 0.84. AUC was 0.85. This study presents the first spatial analysis based on unsupervised machine learning algorithms. Without expert annotation, CluSA framework can not only accurately classify and predict the degree of kidney function at the biopsy and in one year, but also identify novel predictors of kidney function and renal prognosis.

Role of endogenous adenine in kidney failure and mortality with diabetes.

Diabetic kidney disease (DKD) can lead to end-stage kidney disease (ESKD) and mortality, however, few mechanistic biomarkers are available for high risk patients, especially those without macroalbuminuria. Urine from participants with diabetes from Chronic Renal Insufficiency Cohort (CRIC), Singapore Study of Macro-Angiopathy and Reactivity in Type 2 Diabetes (SMART2D), and the Pima Indian Study determined if urine adenine/creatinine ratio (UAdCR) could be a mechanistic biomarker for ESKD. ESKD and mortality were associated with the highest UAdCR tertile in CRIC (HR 1.57, 1.18, 2.10) and SMART2D (HR 1.77, 1.00, 3.12). ESKD was associated with the highest UAdCR tertile in patients without macroalbuminuria in CRIC (HR 2.36, 1.26, 4.39), SMART2D (HR 2.39, 1.08, 5.29), and Pima Indian study (HR 4.57, CI 1.37-13.34). Empagliflozin lowered UAdCR in non-macroalbuminuric participants. Spatial metabolomics localized adenine to kidney pathology and transcriptomics identified ribonucleoprotein biogenesis as a top pathway in proximal tubules of patients without macroalbuminuria, implicating mammalian target of rapamycin (mTOR). Adenine stimulated matrix in tubular cells via mTOR and stimulated mTOR in mouse kidneys. A specific inhibitor of adenine production was found to reduce kidney hypertrophy and kidney injury in diabetic mice. We propose that endogenous adenine may be a causative factor in DKD.

Defining the molecular correlate of arteriolar hyalinosis in kidney disease progression by integration of single cell transcriptomic analysis and pathology scoring.

Arteriolar hyalinosis in kidneys is an independent predictor of cardiovascular disease, the main cause of mortality in chronic kidney disease (CKD). The underlying molecular mechanisms of protein accumulation in the subendothelial space are not well understood. Using single cell transcriptomic data and whole slide images from kidney biopsies of patients with CKD and acute kidney injury in the Kidney Precision Medicine Project, the molecular signals associated with arteriolar hyalinosis were evaluated. Co-expression network analysis of the endothelial genes yielded three gene set modules as significantly associated with arteriolar hyalinosis. Pathway analysis of these modules showed enrichment of transforming growth factor beta / bone morphogenetic protein (TGFß / BMP) and vascular endothelial growth factor (VEGF) signaling pathways in the endothelial cell signatures. Ligand-receptor analysis identified multiple integrins and cell adhesion receptors as over-expressed in arteriolar hyalinosis, suggesting a potential role of integrin-mediated TGFß signaling. Further analysis of arteriolar hyalinosis associated endothelial module genes identified focal segmental glomerular sclerosis as an enriched term. On validation in gene expression profiles from the Nephrotic Syndrome Study Network cohort, one of the three modules was significantly associated with the composite endpoint (> 40% reduction in estimated glomerular filtration rate (eGFR) or kidney failure) independent of age, sex, race, and baseline eGFR, suggesting poor prognosis with elevated expression of genes in this module. Thus, integration of structural and single cell molecular features yielded biologically relevant gene sets, signaling pathways and ligand-receptor interactions, underlying arteriolar hyalinosis and putative targets for therapeutic intervention.

An atlas of healthy and injured cell states and niches in the human kidney.

Understanding kidney disease relies on defining the complexity of cell types and states, their associated molecular profiles and interactions within tissue neighbourhoods1. Here we applied multiple single-cell and single-nucleus assays (>400,000 nuclei or cells) and spatial imaging technologies to a broad spectrum of healthy reference kidneys (45 donors) and diseased kidneys (48 patients). This has provided a high-resolution cellular atlas of 51 main cell types, which include rare and previously undescribed cell populations. The multi-omic approach provides detailed transcriptomic profiles, regulatory factors and spatial localizations spanning the entire kidney. We also define 28 cellular states across nephron segments and interstitium that were altered in kidney injury, encompassing cycling, adaptive (successful or maladaptive repair), transitioning and degenerative states. Molecular signatures permitted the localization of these states within injury neighbourhoods using spatial transcriptomics, while large-scale 3D imaging analysis (around 1.2 million neighbourhoods) provided corresponding linkages to active immune responses. These analyses defined biological pathways that are relevant to injury time-course and niches, including signatures underlying epithelial repair that predicted maladaptive states associated with a decline in kidney function. This integrated multimodal spatial cell atlas of healthy and diseased human kidneys represents a comprehensive benchmark of cellular states, neighbourhoods, outcome-associated signatures and publicly available interactive visualizations.

The chromatin landscape of healthy and injured cell types in the human kidney.

There is a need to define regions of gene activation or repression that control human kidney cells in states of health, injury, and repair to understand the molecular pathogenesis of kidney disease and design therapeutic strategies. However, comprehensive integration of gene expression with epigenetic features that define regulatory elements remains a significant challenge. We measured dual single nucleus RNA expression and chromatin accessibility, DNA methylation, and H3K27ac, H3K4me1, H3K4me3, and H3K27me3 histone modifications to decipher the chromatin landscape and gene regulation of the kidney in reference and adaptive injury states. We established a comprehensive and spatially-anchored epigenomic atlas to define the kidney's active, silent, and regulatory accessible chromatin regions across the genome. Using this atlas, we noted distinct control of adaptive injury in different epithelial cell types. A proximal tubule cell transcription factor network of ELF3 , KLF6 , and KLF10 regulated the transition between health and injury, while in thick ascending limb cells this transition was regulated by NR2F1 . Further, combined perturbation of ELF3 , KLF6 , and KLF10 distinguished two adaptive proximal tubular cell subtypes, one of which manifested a repair trajectory after knockout. This atlas will serve as a foundation to facilitate targeted cell-specific therapeutics by reprogramming gene regulatory networks.