SCODiA Report by MLBI Lab

Pancreatic Cellular Landscape in Type 1 Diabetes: Insights from Single-Cell Analysis of Cell Population Dynamics, Interactions, and Pathway Perturbations

This single-cell RNA-seq analysis of human pancreas tissue reveals profound cellular and molecular changes associated with Type 1 Diabetes (T1D) and autoantibody-positive states. Key findings include significant beta cell depletion and a prominent shift towards pro-inflammatory M1 macrophage polarization in overt T1D. Furthermore, cell-cell interaction analysis identified altered signaling between exocrine and stromal cells, while gene set enrichment showed widespread inflammatory, stress, and metabolic dysregulation across multiple pancreatic cell types, particularly beta cells. These insights highlight critical pathogenic mechanisms and potential therapeutic avenues for T1D and its early stages.

Contents

0. Dataset overview
1. UMAP Visualization of Pancreatic Single-Cell RNA-Seq Data by Condition, Sample, and Cell Type Annotation
2. Cell Type Annotation Assessment via UMAP Projection of Major Cell Type Scores
3. Overall Celltype_subset Marker Gene Expression Dot Plot in Pancreas
4. Pancreatic Minor Cell Type Population Analysis Across Disease Conditions
5. Macrophage Subset Composition Across Pancreatic Conditions
6. Cell-Cell Interaction Analysis Across Pancreatic Conditions
7. Macrophage Condition-Specific Surfaceome Markers in Pancreas
8. Fibroblast Condition-Specific Surfaceome Markers in Type 1 Diabetes
9. Gene Ontology (GSA) Analysis of Upregulated Pathways in Pancreatic Cell Types Across Diabetic Conditions
10. Gene Set Enrichment Analysis of Pancreatic Cell Types Across Diabetes Conditions
11. Discussion
12. Query List

0. Dataset overview

Dataset Summary

• This dataset contains single-cell RNA-seq data from human pancreas, pre-processed by SCODA.
• It includes data for 40029 cells and 30501 genes.
• Key metadata (obs columns) available for each cell include sample, donor, condition (type1_diabetes, non_diabetic, autoantibody_positive), sex, n_counts, n_genes, pct_mito, and various cell type annotations: celltype_major, celltype_minor, celltype_subset, and cluster.
• The reference condition for differential analyses is non_diabetic.

Precomputed results are available for

• Cell-Cell Interactions (CCI) per condition and sample.
• Differential Gene Expression (DEG) comparing conditions (versus rest or versus reference) for various cell types.
• Gene Set Enrichment Analysis (GSEA) per cell type, comparing conditions (versus rest or versus reference).
• Gene Ontology (GO/GSA) enrichment per cell type, comparing conditions (versus rest or versus reference).
• For DEG, GSEA, and GSA/GO analyses, target cell types (minor level) include: Acinar cell, Alpha cell, Beta cell, Ductal cell, Fibroblast, Macrophage, Stellate cell.

1. UMAP Visualization of Pancreatic Single-Cell RNA-Seq Data by Condition, Sample, and Cell Type Annotation

[Analysis Visualization Results]...

Analysis Overview

This analysis presents a Uniform Manifold Approximation and Projection (UMAP) visualization of single-cell RNA-seq data from the human pancreas, comprising over 40,000 cells and 30,000 genes. The UMAP plots are colored by various metadata attributes: condition (autoantibody_positive, non_diabetic, type1_diabetes), sample (individual donors), celltype_major, celltype_minor, and celltype_subset. These visualizations provide an essential initial overview of the dataset structure, cell type clustering, and the distribution of samples and conditions across the cellular landscape. This helps assess data quality, annotation consistency, and potential biological differences related to disease conditions.

Visual Summary

Embedding Structure

The UMAP displays a complex, multi-lobed structure, characteristic of single-cell data with diverse cell populations. There is a large, central cluster, from which several distinct smaller clusters and "arms" emanate. This suggests the presence of a few highly abundant cell types alongside several more specialized or rarer populations.

condition Distribution

The UMAP colored by condition shows that cells from "non_diabetic" (yellow) are broadly distributed across the main body of the UMAP, forming a foundational layer. "type1_diabetes" (dark blue) cells are also widespread but appear to be somewhat enriched in certain regions, particularly in smaller peripheral clusters. "autoantibody_positive" (maroon) cells seem to be present in all major regions but do not dominate any particular cluster or show strong segregation, indicating a mixed population that might represent an early or transitional disease state. Overall, there's no single large cluster uniquely dominated by any one condition, suggesting that major cell type populations are present across conditions, though their proportions or states might differ.

sample Distribution

The sample UMAP reveals a relatively good mixing of cells from different donors across the majority of the embedding space. Cells from various 'MM_xxx' samples (represented by different colors) are largely interspersed within most clusters, especially the central dense cluster. This intermixing is a positive indicator, suggesting that potential batch effects originating from individual samples are not strongly driving the major cellular separation in the UMAP. However, upon closer inspection, some smaller, more distinct clusters at the periphery show a slight enrichment of cells from particular samples, which could warrant further investigation.

celltype_major Distribution

The celltype_major plot clearly delineates the primary cell populations. "Acinar cells" (maroon) form the largest and most dense cluster, consistent with their abundance in the exocrine pancreas. "Ductal cells" (light yellow) constitute another large cluster, distinct from Acinar cells but broadly distributed. Islet cells, represented by "Alpha cells" (orange) and "Beta cells" (red), form well-separated clusters, primarily in one of the "arms" emanating from the main body, indicating their distinct transcriptional profiles. "Myeloid cells" (light green) and "Stromal cells" (dark blue) also form distinct, smaller clusters, consistent with their presence as immune and supportive cells in the pancreatic tissue. The clear separation of these major cell types underscores the quality of the cell type annotation.

celltype_minor Distribution

Refining the celltype_major view, the celltype_minor plot provides more granular cell identities. The main "Acinar cell" and "Ductal cell" populations remain prominent. Within the stromal compartment, "Fibroblast" (light yellow) and "Stellate cell" (teal) populations are now visible as distinct clusters. "Macrophage" (light green) and "DC" (orange) cells are resolved from the broader "Myeloid cell" category, further confirming the presence of distinct immune cell populations. "Alpha cells" and "Beta cells" remain distinct, as expected for their critical roles in islet function. The presence of an "unassigned" cluster (dark blue) indicates a small population of cells that could not be confidently assigned to a known minor cell type, warranting further investigation.

celltype_subset Distribution

The celltype_subset UMAP provides the most detailed cell type resolution. This plot shows further sub-clustering within the minor cell types, particularly for immune cells. For example, "Macrophage" cells are further divided into subsets like "Mac_M1," "Mac_M2A," "Mac_M2B," "Mac_M2C," and "Mac_M2D," which exhibit some degree of spatial separation within the broader Macrophage cluster. Similarly, "DC" cells are refined into "cDC," "iDC," and "pDC" subsets. This granular annotation is crucial for understanding the immune microenvironment in the context of diabetes. The "unassigned" population persists as a distinct cluster.

Biological Interpretation

The UMAP plots collectively demonstrate a robust and well-annotated single-cell dataset of the human pancreas.

1. Pancreatic Cell Type Architecture: The clear separation of major pancreatic cell types (Acinar, Ductal, Alpha, Beta) and supporting stromal and immune cells (Fibroblast, Stellate, Macrophage, DC) indicates successful capture and annotation of the cellular diversity within the pancreas. The distinct clustering of Alpha and Beta cells, which are crucial for glucose homeostasis, suggests their unique gene expression profiles are well-resolved.
2. Immune Cell Heterogeneity: The progressive refinement from celltype_major to celltype_subset highlights the complex heterogeneity of immune cells, particularly macrophages and dendritic cells. The identification of various Macrophage and DC subsets (e.g., M1, M2 subtypes for Macrophages, and classical, inflammatory, plasmacytoid DCs) is significant given the known role of immune cell infiltration and activation in the pathogenesis of type 1 diabetes (T1D) [PubMed Search]. Different macrophage polarization states (M1 vs. M2) are associated with pro-inflammatory and anti-inflammatory/reparative roles, respectively, which could be critical in understanding disease progression.
3. Disease Impact on Cell Populations: While no single large cluster is exclusively driven by a disease condition, the enriched presence of "type1_diabetes" cells in some peripheral clusters, especially those containing immune cells, suggests that T1D might be associated with shifts in specific immune cell populations or activation states. Similarly, the "autoantibody_positive" condition, representing individuals at risk or in early stages of T1D, shows a mixed distribution, which could reflect the ongoing immunological processes before full-blown disease manifestation. This pattern implies that T1D pathology might involve subtle changes in the composition or state of existing cell types rather than the emergence of entirely novel populations.
4. Data Quality: The good intermixing of sample across most clusters suggests that biological variability (cell type differences) rather than technical batch effects is the primary driver of the UMAP embedding. This enhances confidence in downstream differential expression or cell-cell interaction analyses.
5. Unassigned Cells: The presence of an "unassigned" celltype_minor and celltype_subset population, while small, indicates that there might be rare or novel cell types whose identities are yet to be fully characterized, or perhaps cells with ambiguous transcriptional profiles. Further investigation into these cells could uncover new biological insights.

Clinical or Translational Implications

The fine-grained cell type resolution, especially within the immune compartment, offers valuable insights for understanding the immunological processes in T1D. Identifying specific macrophage or DC subsets that are enriched or display altered states in "type1_diabetes" or "autoantibody_positive" conditions could point to critical immune cell players in disease initiation and progression. This understanding could inform the development of targeted immunotherapies or biomarkers for early diagnosis and disease monitoring in T1D. For instance, if a specific pro-inflammatory macrophage subset (e.g., M1-like) is found to be significantly expanded in T1D, it could be a therapeutic target. Conversely, understanding the distribution of "autoantibody_positive" cells could help characterize the immune landscape in individuals at high risk for developing T1D, potentially revealing early immunological signatures of disease.

Annotation Notes

The comprehensive and hierarchical cell type annotation (major, minor, subset) is a strength of this dataset, enabling detailed investigation of pancreatic cell biology. The distinct separation of most annotated cell types in the UMAPs indicates high confidence in the current annotations. The presence of a small "unassigned" population suggests an area for potential future refinement, possibly through integration with additional markers or reference atlases.

2. Cell Type Annotation Assessment via UMAP Projection of Major Cell Type Scores

[Analysis Visualization Results]...

Analysis Overview

This analysis visualizes the distribution of major cell type scores across a Uniform Manifold Approximation and Projection (UMAP) embedding of single-cell RNA-seq data from the human pancreas. Each plot displays the score for a specific major cell type, indicating the confidence of a cell belonging to that type, with higher scores (yellow/green) signifying stronger classification. The final plot shows the assigned celltype_major annotation, providing an overall view of the cell type clustering. The purpose is to assess the quality and spatial segregation of the annotated major cell populations within the dataset.

Visual Summary

The UMAP projection organizes the 40,029 cells into a distinct, branched structure, effectively separating various pancreatic cell populations.

• Overall UMAP Structure: The UMAP resembles a hand-like structure, with a large central "palm" region and several "finger-like" extensions.

Major Cell Type Clustering:

• Acinar cells form the most prominent and dense cluster, occupying the large "palm" region of the UMAP, with very high Acinar cell scores localized precisely to this area.
• Ductal cells form a distinct, well-separated "finger" cluster, showing high Ductal cell scores in this specific region.
• Endocrine Islet Cells (Alpha, Beta, Delta, Gamma (PP), Epsilon cells): These cell types collectively form several interconnected "finger" clusters. Alpha and Beta cells, the two most abundant endocrine populations, each occupy distinct, high-scoring regions within these clusters. Delta, Gamma (PP), and Epsilon cells, representing rarer endocrine populations, also show highly localized scores within their respective, smaller clusters, indicating clear separation.

Stromal and Immune Cells:

• Stromal cells form another well-defined cluster, clearly separated from both exocrine and endocrine compartments, with high Stromal cell scores concentrated there. This broader stromal compartment likely includes various mesenchymal cell types (Fibroblasts, Stellate cells) and potentially Endothelial cells.
• Endothelial cells show a distinct, albeit smaller, cluster within or adjacent to the stromal compartment.
• Myeloid cells, T cells, B cells, and Mast cells all form discrete, relatively smaller clusters, primarily located in distinct "finger" regions, indicating well-resolved immune cell populations. Myeloid cell scores precisely overlap with the celltype_major annotation for Myeloid cells.

Rare/Specialized Cells:

• Pancreatic progenitor cells show a somewhat more diffuse pattern but with enriched scores in specific areas, potentially indicating transitional states or lower abundance.
• Schwann cells form a very small, distinct cluster, highlighting the presence of neural components.
• Consistency with celltype_major: The individual HiCAT major cell type score plots show excellent concordance with the celltype_major annotation. Regions with high scores for a particular cell type precisely align with the assigned clusters for that cell type in the celltype_major plot, indicating robust and consistent cell identity assignments.

Biological Interpretation

The UMAP projection, colored by major cell type scores, provides a clear and robust visualization of the cellular landscape of the human pancreas.

• Distinct Pancreatic Compartments: The analysis clearly delineates the major functional compartments of the pancreas: the exocrine parenchyma (dominated by Acinar cells, with Ductal cells forming the ductal tree), the endocrine islets (comprising Alpha, Beta, Delta, Gamma, and Epsilon cells), and the supporting stromal and immune microenvironment. The physical separation of these populations on the UMAP reflects their distinct transcriptional programs and specialized functions.
• High-Resolution Islet Cell Identification: The individual score plots for Alpha, Beta, Delta, Gamma, and Epsilon cells confirm the presence and distinct transcriptional profiles of these critical endocrine cell types. This is crucial for studying pancreatic physiology and diseases like diabetes, where specific islet cell populations are affected GeneCards: Insulin (Beta cell marker), GeneCards: Glucagon (Alpha cell marker).
• Immune and Stromal Cell Context: The identification of various immune cell subsets (T, B, Myeloid, Mast cells) and diverse stromal cells (Stromal, Endothelial, Schwann) highlights the complex cellular milieu of the pancreas. These non-parenchymal cells play crucial roles in maintaining tissue homeostasis, mediating inflammatory responses, and contributing to disease pathogenesis, particularly in conditions like Type 1 Diabetes and pancreatic cancer PubMed search: Pancreatic immune microenvironment.
• Annotation Quality: The highly localized and strong cell type scores for most populations indicate a high degree of confidence in the automated annotation process (HiCAT) and the underlying single-cell transcriptomic data quality. This robust annotation forms a solid foundation for subsequent differential gene expression, pathway analysis, and cell-cell interaction studies.

Annotation Notes

The UMAP visualization of major cell type scores confirms the high quality and specificity of the cell type annotations for this pancreatic single-cell RNA-seq dataset.

• Strong Cell Identity: The clear clustering and distinct localization of high scores for each major cell type demonstrate strong cell identity based on transcriptional profiles. This suggests that the chosen cell type markers and classification approach are highly effective.
• Well-Resolved Populations: The separation of closely related cell types, especially within the endocrine compartment (e.g., Alpha vs. Beta cells), and the distinct identification of minor populations like Pancreatic progenitor cells and Schwann cells, underscore the excellent resolution of the dataset and annotation.
• Foundation for Downstream Analysis: The robust and well-validated cell type annotations provided by this analysis are essential for accurate interpretation of any subsequent differential expression, pathway enrichment, or cell-cell interaction analyses performed on this dataset.

3. Overall Celltype_subset Marker Gene Expression Dot Plot in Pancreas

[Analysis Visualization Results]...

Analysis Overview

This analysis generates a dot plot visualizing the expression patterns of marker genes across different celltype_subset populations identified in the single-cell RNA-seq data from human pancreas tissue. The plot serves as a crucial quality control step for cell type annotation, demonstrating whether the assigned cell identities are supported by the expression of known, specific marker genes. The dot size represents the fraction of cells within each group expressing a particular gene, while the color intensity indicates the mean expression level of that gene within the group. The markers shown were identified with a focus on surfaceome proteins, though some highly expressed intracellular markers may also be present to support robust cell identification.

Visual Summary

The dot plot effectively visualizes the specificity and expression levels of key marker genes for each celltype_subset.

• Distinct Expression Profiles: The plot shows clear blocks of highly expressed genes (large, dark red dots) specific to individual cell types, confirming the distinct transcriptional identities of the annotated cell populations. Red boxes highlight these specific marker gene clusters for each cell type.
• Fraction of Expressing Cells: The size of the dots indicates that most of the identified markers are expressed in a high fraction of cells within their respective celltype_subset groups, further supporting their utility as robust markers.
• Mean Expression Levels: The intensity of the red color signifies high mean expression of these markers within their target cell types.
• Cell Type Abundance: The bar chart on the right indicates the total number of cells for each celltype_subset. Acinar cells are the most abundant cell type (30,739 cells), followed by Alpha cells (1,044 cells) and Beta cells (926 cells), reflecting the composition of pancreatic tissue.
• Specificity of Markers: Most markers exhibit high specificity, with strong expression confined to one or a few closely related cell types. For example, insulin (INS) is highly specific to Beta cells, and glucagon (GCG) to Alpha cells.
• Overlap in Stromal Cells: Some overlap in marker gene expression is observed between Fibroblasts and Stellate cells (e.g., COL1A1, DCN). This is biologically plausible, as both are mesenchymal cell types and pancreatic stellate cells can adopt myofibroblast-like characteristics upon activation.

Biological Interpretation

The marker gene expression patterns strongly support the assigned celltype_subset annotations and provide valuable insights into the cellular composition of the human pancreas.

• Acinar Cells: These cells are robustly identified by the expression of numerous digestive enzyme genes and zymogen granule proteins, including PNLIPRP1, CPA2, PRSS1, CEL, CTRB1, SPINK1, and CLPS (pancreatic lipase-related protein 1, carboxypeptidase A2, trypsinogen 1, carboxyl ester lipase, chymotrypsin B1, serine protease inhibitor Kazal type 1, carboxyl ester lipase). These genes are essential for exocrine pancreatic function, confirming the identity of acinar cells [GeneCards].
• Alpha Cells: Distinctive expression of GCG (glucagon), a key hormone regulating blood glucose levels, along with transcription factors like ARX and MAFB, confirms their identity as pancreatic alpha cells, responsible for glucagon secretion [UniProt].
• Beta Cells: These cells are clearly defined by the high expression of INS (insulin) and IAPP (amylin), the primary hormones of the endocrine pancreas, alongside critical transcription factors like PDX1 and NKX6-1, and glucose transporter SLC2A2 (GLUT2). These markers are crucial for glucose-stimulated insulin secretion and beta cell identity [GeneCards].
• DC (Classical): Classical Dendritic cells are characterized by markers such as CD1C and HLA-DRA, which are involved in antigen presentation and immune activation, consistent with their role in immune surveillance [GeneCards].
• Ductal Cells: Identified by KRT7 (keratin 7), CFTR (cystic fibrosis transmembrane conductance regulator), and tight junction proteins like CLDN1 and CLDN3 (claudin 1 and 3), these markers align with the structural and transport functions of pancreatic ductal epithelium [GeneCards].
• Fibroblasts: These stromal cells show high expression of extracellular matrix components and related proteins, including various collagen genes (COL1A1, COL3A1, COL5A1, COL6A1), DCN (decorin), FAP (fibroblast activation protein), and LUM (lumican), reflecting their role in tissue structure and remodeling [GeneCards].

Macrophage Subsets (M1, M2A, M2C):

• Macrophage (M1): Identified by IRF5, CCL2, and CD80, consistent with a pro-inflammatory profile often associated with pathogen defense [PubMed Search].
• Macrophage (M2A): Characterized by markers like IL1R1, CLEC7A, and CD36, indicating roles in allergic responses, parasitic infections, and tissue repair [UniProt].
• Macrophage (M2C): Distinguished by MSR1, IL1R2, TLR1, and PPARG, suggesting functions in immunosuppression, tissue remodeling, and efferocytosis [GeneCards].
• Stellate Cells: These cells express ACTA2 (alpha-smooth muscle actin), DCN (decorin), and collagen genes like COL1A1. The expression of ACTA2 is a hallmark of activated stellate cells (myofibroblasts), which play a critical role in pancreatic fibrosis. Their shared expression of DCN and COL1A1 with fibroblasts underscores the continuum and potential interconversion between these mesenchymal cell types in the pancreatic stroma [GeneCards].

Annotation Notes

The comprehensive display of highly specific marker gene expression patterns provides strong evidence for the accuracy and robustness of the celltype_subset annotations in this dataset. The presence of well-established markers for each cell type, with distinct and high expression, confirms their identity. The clear separation of endocrine islet cells (Alpha, Beta), exocrine (Acinar, Ductal), stromal (Fibroblast, Stellate), and immune cells (DC, Macrophage subsets) based on their gene expression profiles is evident.

It is worth noting that while the find_cfg parameter specified surfaceome_only: True for marker identification, some highly classical and specific intracellular/secreted markers (e.g., INS, GCG, PRSS1) are present in the plot. This suggests that the marker selection logic may prioritize canonical cell identity markers even if they are not strictly surface proteins, or that the "surfaceome" definition used might be broad. However, the overall result strongly supports the biological validity of the cell type assignments. The high number of markers plotted (up to 30 per group) further enhances confidence in the specific identities of these pancreatic cell populations.

4. Pancreatic Minor Cell Type Population Analysis Across Disease Conditions

[Analysis Visualization Results]...

Analysis Overview

This analysis presents the relative proportions of minor cell types within individual pancreatic samples, stratified by disease condition: 'autoantibody_positive', 'non_diabetic', and 'type1_diabetes'. The stacked bar plots provide a visual summary of the cellular composition of the pancreatic tissue derived from single-cell RNA sequencing data for each sample. This type of visualization is crucial for understanding baseline cellular heterogeneity and how it shifts in different disease states, particularly in conditions like Type 1 Diabetes (T1D) where specific cell populations, such as beta cells, are targeted.

Visual Summary

The stacked bar plots display the relative abundance of nine minor cell types across various samples for each of the three conditions.

• Dominant Cell Types: Acinar cells (dark red) consistently represent the largest proportion of cells across almost all samples and conditions, often exceeding 60-80% of the total cells. Ductal cells (light yellow) are also a significant component, typically making up 10-30%. This dominance of exocrine cells (Acinar and Ductal) is expected in pancreatic tissue.

Beta Cell Proportions (Orange)

• In non_diabetic samples, Beta cells (orange) are consistently present, typically forming a visible, albeit small, proportion (around 1-5% in most samples).
• In autoantibody_positive samples, Beta cells are also present, similar to non-diabetic controls, indicating that these individuals, despite having autoantibodies, may not have experienced significant Beta cell loss yet.
• In type1_diabetes samples, Beta cell populations appear noticeably reduced or almost absent in many individuals (e.g., MM_401, MM_555, MM_406, MM_398, MM_335). The remaining Beta cell signal is minimal in several T1D samples, indicating severe depletion.
• Alpha Cell Proportions (Red): Alpha cells (red), another endocrine cell type, are present across all conditions. While their proportions are generally smaller than Ductal cells, they appear relatively more stable or less severely affected compared to Beta cells in Type 1 Diabetes, which is a known aspect of T1D progression where alpha cells can be more resistant or even expand relatively PubMed Link.
• Immune Cell Populations (Macrophage, Dendritic cell): Macrophages (pale yellow/light green) and Dendritic cells (light orange) are present in very small proportions across all conditions. While their relative abundance is low, their presence is notable in the context of an autoimmune disease. Clear significant increases in these cell types are not overtly visible from this relative population plot, but their persistence suggests a baseline immune presence.
• Stromal Cell Populations (Fibroblast, Stellate cell, Smooth muscle cell): Fibroblasts (cream yellow), Stellate cells (teal), and Smooth muscle cells (light green) also constitute minor proportions. No dramatic shifts in these stromal components are immediately evident across conditions.
• Sample Heterogeneity: There is noticeable sample-to-sample variability within each condition regarding the exact proportions of each cell type, highlighting inter-individual differences in cellular composition.

Biological Interpretation

The observed shifts in minor cell type populations provide critical biological insights into the pathology of Type 1 Diabetes:

• Beta Cell Depletion in T1D: The most striking observation is the profound reduction or near absence of Beta cells in many samples from individuals with type1_diabetes. This finding strongly aligns with the known autoimmune destruction of insulin-producing pancreatic Beta cells, which is the hallmark of T1D pathogenesis GeneCards: INS gene. This relative depletion of Beta cells results in a higher relative proportion of other surviving cell types, such as Acinar cells.
• Alpha Cell Resilience/Compensation: Alpha cells, which produce glucagon, appear relatively more preserved than Beta cells in T1D. This differential vulnerability of islet cell types is a known feature of the disease, and understanding the mechanisms behind it could offer therapeutic avenues PubMed Link.
• Early Disease Stage (Autoantibody Positive): Individuals classified as autoantibody_positive do not show a significant reduction in Beta cell populations compared to non_diabetic individuals in this dataset. This suggests that at the point of sample collection, overt Beta cell destruction might not have progressed to the point of major population shifts, or that Beta cell loss is highly heterogeneous in this pre-diabetic stage. These individuals are at high risk for developing T1D, making them critical for studying early disease mechanisms PubMed Link.
• Immune Cell Infiltration: The consistent, albeit low, presence of Macrophages and Dendritic cells across all conditions, including autoantibody-positive and T1D samples, is biologically relevant. In T1D, these immune cells play a crucial role in initiating and perpetuating the autoimmune destruction of Beta cells (insulitis) PubMed Link. While a significant increase isn't visually obvious from this relative plot, their presence provides context for potential future differential expression or cell-cell interaction analyses.
• Exocrine and Ductal Cell Dominance: The consistent high proportion of Acinar and Ductal cells reflects their fundamental role as the primary structural and functional units of the exocrine pancreas. Their relative increase in T1D samples is likely a consequence of the relative loss of endocrine (islet) cells, rather than an absolute expansion.

Clinical or Translational Implications

• Biomarker for T1D Progression: The marked reduction in Beta cell proportions, as observed in this scRNA-seq analysis, serves as a molecular confirmation of the pathological changes in Type 1 Diabetes. Quantifying Beta cell mass using advanced imaging or cellular profiling could potentially serve as a biomarker for disease severity and progression.
• Understanding Pre-symptomatic Stages: The similar Beta cell proportions in autoantibody_positive individuals to non_diabetic controls highlights the challenge of identifying early pancreatic changes during the pre-symptomatic phase of T1D. Future analyses should focus on subtle molecular shifts (e.g., differential gene expression or cellular stress markers) within specific cell types in autoantibody-positive individuals, even if their gross population numbers are unchanged.
• Targeting Immune and Stromal Interactions: The presence of immune cells (Macrophages, Dendritic cells) even in low proportions, especially in disease contexts, points to the potential importance of targeting immune cell function or their interactions with islet and stromal cells to halt or slow disease progression.
• Heterogeneity in Disease: The sample-to-sample variability observed underscores the importance of personalized approaches in T1D research and treatment. Different patients may exhibit varying degrees of Beta cell loss and immune infiltration, necessitating tailored therapeutic strategies.

5. Macrophage Subset Composition Across Pancreatic Conditions

[Analysis Visualization Results]...

Analysis Overview

This analysis presents a stacked bar plot visualizing the relative proportions of various macrophage subsets (M1, M2A, M2B, M2C, M2D) within the total macrophage population for individual samples across three conditions: autoantibody_positive, non_diabetic (reference), and type1_diabetes. The data is derived from single-cell RNA sequencing of human pancreas samples, allowing for a detailed examination of macrophage heterogeneity in the context of Type 1 Diabetes (T1D).

Visual Summary

The visualization clearly illustrates distinct patterns of macrophage subset distribution across the different conditions:

• Type 1 Diabetes (type1_diabetes): Samples from individuals with overt Type 1 Diabetes show a striking and consistent dominance of Macrophage (M1) (burgundy). In many type1_diabetes samples, M1 macrophages constitute a very high proportion, often exceeding 80-90% of the total macrophage population, and in one case (MM_355), nearly 100%. Conversely, the proportions of M2 subsets (M2A, M2B, M2C, M2D) are significantly reduced or almost absent in these samples.
• Non-Diabetic (non_diabetic): The non_diabetic control samples exhibit a more diverse macrophage composition. While M1 macrophages are present, they typically represent a lower proportion compared to the type1_diabetes group. Macrophage (M2C) (pale yellow) is particularly prominent in non_diabetic samples, often making up a substantial part of the total macrophage pool, suggesting a more homeostatic or regulatory environment. Other M2 subsets (M2A, M2B, M2D) are also observed in varying, though generally lower, proportions.
• Autoantibody Positive (autoantibody_positive): This group, representing individuals at high risk for T1D, shows an intermediate and somewhat heterogeneous macrophage profile. While some autoantibody_positive samples exhibit a composition closer to non_diabetic (e.g., higher M2C), others show an increased proportion of M1 macrophages compared to controls, though generally not as pronounced as in overt type1_diabetes. The overall trend suggests a shift towards increased M1 polarization as compared to non_diabetic controls, with considerable sample-to-sample variability.

In summary, there is a clear shift towards an M1-dominant macrophage phenotype in overt Type 1 Diabetes, with the autoantibody_positive group showing an evolving pattern between non-diabetic controls and full disease.

Biological Interpretation

The observed macrophage subset repartitioning provides critical insights into the inflammatory processes in the pancreas related to Type 1 Diabetes:

• M1 Macrophages and Pro-inflammatory Role: M1 macrophages are classically characterized as pro-inflammatory, producing cytokines such as TNF-α, IL-1β, and IL-6. Their significant enrichment in the pancreas of type1_diabetes patients strongly suggests a highly inflammatory environment contributing to the destruction of insulin-producing beta cells, a hallmark of T1D pathogenesis. This sustained M1-driven inflammation likely fuels insulitis. [PubMed search: M1 macrophages type 1 diabetes]
• M2 Macrophages and Immunomodulation/Repair: M2 macrophages, encompassing various subtypes like M2A, M2B, M2C, and M2D, are generally associated with anti-inflammatory responses, immune regulation, tissue repair, and resolution of inflammation. The relative abundance of M2C macrophages in non_diabetic individuals suggests their role in maintaining pancreatic homeostasis and resolving any low-level inflammation. The sharp decline of M2 populations, particularly M2C, in type1_diabetes patients indicates a failure of these regulatory mechanisms and an unchecked inflammatory state. [GeneCards: CD163 (M2 macrophage marker)]
• Progression of Autoimmunity: The autoantibody_positive group, representing a pre-symptomatic stage of T1D, displays an evolving macrophage landscape. The mixed profile, with some samples showing M1 increases, suggests that the pancreatic immune environment may already be undergoing shifts towards pro-inflammatory states even before the clinical onset of diabetes. This heterogeneity could reflect different stages of progression or individual variations in the autoimmune response.

Clinical or Translational Implications

The distinct macrophage polarization patterns identified in this analysis hold several clinical and translational implications:

• Biomarker Potential: The elevated proportion of M1 macrophages, particularly relative to M2 subsets, could serve as a valuable biomarker for T1D progression or disease activity within the pancreas. Monitoring this ratio might offer insights into disease staging or response to immunomodulatory therapies.
• Therapeutic Targets: Modulating macrophage polarization could be a promising therapeutic strategy for T1D. Interventions aimed at shifting the M1-dominant phenotype towards a more M2-like, anti-inflammatory or regulatory state in the pancreas could potentially preserve beta cell function, slow disease progression, or even prevent the onset of clinical T1D in at-risk autoantibody_positive individuals. Strategies might include targeting specific signaling pathways that drive M1 polarization or promoting M2 differentiation. [PubMed search: macrophage polarization therapy type 1 diabetes]
• Understanding Insulitis: These findings provide further evidence for the critical role of macrophages in the pathogenesis of T1D-associated insulitis, where immune cells infiltrate and damage pancreatic islets. A detailed understanding of macrophage subsets and their functions in the pancreatic microenvironment is crucial for developing targeted immune therapies.

6. Cell-Cell Interaction Analysis Across Pancreatic Conditions

[Analysis Visualization Results]...

Analysis Overview

This analysis investigates cell-cell interactions (CCI) within the human pancreas, focusing on interactions between Ductal and Acinar cells, across three distinct conditions: 'non_diabetic' (reference), 'autoantibody_positive' (pre-diabetic state), and 'type1_diabetes' (frank disease state). The plot_cci_dots tool was utilized to visualize the top 80 significant ligand-receptor pairs for each condition, based on their p-values and mean expression levels. The goal is to identify common and condition-specific intercellular communication patterns that may shed light on disease pathogenesis in Type 1 Diabetes (T1D).

Visual Summary

The three dot plots visually represent the most significant cell-cell interactions. Each plot displays cell-pair interactions (e.g., Ductal-Ductal, Ductal-Acinar, Acinar-Ductal, Acinar-Acinar) on the y-axis and specific ligand-receptor gene pairs on the x-axis. The size of each dot indicates the significance of the interaction (-log10(p-value), larger dots signifying lower p-values and thus higher significance), while the color reflects the mean expression level of the ligand-receptor pair (log2(mean), warmer colors indicating higher expression).

Key observations include:

• Dominant Cell-Cell Pairs: Interactions between Ductal and Acinar cells, in both homologous (Ductal-Ductal, Acinar-Acinar) and heterologous (Ductal-Acinar, Acinar-Ductal) contexts, are prominently featured across all conditions.
• Conserved Interactions: Several ligand-receptor pairs appear consistently across all three conditions, albeit with varying intensity (size and color). Examples include various NECTIN and CADM family interactions, components of the NRG-ERBB signaling, and certain SEMA-PLXNB interactions. This suggests maintenance of fundamental communication pathways in the pancreatic microenvironment.

Condition-Specific Patterns:

• non_diabetic: Shows a broad range of interactions, including prominent signaling through EGF-EGFR, FGF10-FGFR2, TGFB2-TGFbeta receptor, and various cholesterol and glutamate-mediated interactions. These likely represent baseline physiological communication.
• autoantibody_positive: While many interactions overlap with 'non_diabetic', some distinct changes are observed. For instance, specific cholesterol-mediated interactions and glutamate signaling appear altered or highly prominent in this pre-diabetic state.
• type1_diabetes: Exhibits specific shifts. For example, LAMC1_integrin_a6b1_complex interaction becomes more prominent in Acinar-Acinar and Acinar-Ductal cells, suggesting potential changes in cell adhesion or extracellular matrix interactions. Other interactions, like Glutamate_byGLS_and_SLC1A2_GRIK2, which were visible in non_diabetic, seem diminished or absent in type1_diabetes within the top 80 pairs, suggesting altered metabolic communication. EGF_EGFR and FGF10_FGFR2 signaling appear generally strong across all conditions but show nuanced differences in significance and mean expression.

Biological Interpretation

The observed cell-cell interactions highlight critical communication axes between pancreatic Acinar and Ductal cells, which are vital for maintaining pancreatic homeostasis and are implicated in the pathophysiology of T1D.

1. Growth Factor and Developmental Signaling:

• EGF-EGFR and FGF10-FGFR2 signaling are consistently observed. These pathways are crucial for pancreatic development, regeneration, and cell survival. Alterations in their strength could impact the pancreas's ability to cope with stress or injury, or to regenerate in response to autoimmune attack. FGF10, in particular, is a known mitogen for pancreatic epithelial cells and plays a role in ductal morphogenesis and islet regeneration [1, 2].
• NRG-ERBB (Neuregulin-ERBB) interactions are also prevalent. Neuregulins are growth factors that regulate cell proliferation, differentiation, and survival, especially in epithelial tissues. ERBB4, a receptor for Neuregulins, is involved in cell-cell communication and has roles in various biological processes, including immune response modulation [3].
• JAG1-NOTCH signaling, a fundamental pathway for cell fate determination, differentiation, and tissue patterning, is consistently present. Dysregulation of Notch signaling has been implicated in various pancreatic diseases, including diabetes and cancer, as it governs pancreatic progenitor cell maintenance and differentiation [4].

2. Cell Adhesion and Extracellular Matrix (ECM) Interactions:

• NECTIN/CADM family interactions are important for homotypic and heterotypic cell adhesion, influencing tissue structure and integrity.
• The increased prominence of LAMC1_integrin_a6b1_complex in type1_diabetes between Acinar-Acinar and Acinar-Ductal cells is noteworthy. Laminin (LAMC1) and integrins are key components of the ECM and cell-matrix adhesion, respectively. Upregulation of these interactions could indicate remodeling of the pancreatic microenvironment, increased cellular stress, or attempts at tissue repair in the context of T1D progression [5]. Such changes in ECM can also influence immune cell infiltration and survival of resident pancreatic cells.

3. Metabolic and Lipid-Mediated Signaling:

• Interactions involving Cholesterol_byCEL_RORA/RORC and Dehydroepiandrosterone_bySTS_PPARA point to significant lipid-mediated communication. Cholesterol and its metabolites, as well as steroid hormones like DHEA, can act as signaling molecules that regulate gene expression and cellular function, potentially influencing metabolic stress, inflammation, and immune responses in the pancreas [6].
• The presence of Glutamate_byGLS_and_SLC1A2_GRIK2 signaling in non_diabetic suggests a role for glutamate in cell communication, possibly linked to metabolic processes or stress responses, and its apparent reduction in type1_diabetes might indicate altered metabolic states in the diseased pancreas.

4. Immune and Stress Responses:

• TGFB2-TGFbeta receptor signaling is a critical regulator of immune responses, fibrosis, and cell growth. Its consistent presence highlights its importance in maintaining tissue homeostasis and its potential role in modulating inflammation and tissue remodeling in the context of T1D autoimmunity [7].
• SEMA4D-PLXNB1/2 interactions are involved in diverse biological processes, including immune cell migration, axon guidance, and angiogenesis. Their presence suggests roles in directing cellular movements or modulating immune responses within the pancreatic microenvironment.

Clinical or Translational Implications

The identification of condition-specific changes in cell-cell interactions within the pancreas, particularly involving Acinar and Ductal cells, has several translational implications:

1. Early Biomarkers: The altered CCI landscape in the autoantibody_positive state, a precursor to T1D, could offer insights into early disease mechanisms. Specific ligand-receptor pairs whose interactions are significantly changed during this phase could serve as potential biomarkers for predicting T1D progression or identifying individuals at higher risk.
2. Therapeutic Targets: Ligand-receptor pairs that show strong and differential activity in type1_diabetes represent promising therapeutic targets.

• For example, if the LAMC1-integrin axis is aberrantly activated in T1D, targeting this interaction could mitigate pathological ECM remodeling or cell adhesion changes that contribute to disease progression.
• Similarly, modulating growth factor pathways (EGF, FGF, NRG) or NOTCH signaling, if they are found to contribute to detrimental processes like inflammation or impaired regeneration, could offer novel therapeutic avenues.
• Understanding the role of cholesterol-mediated and glutamate-mediated signaling in disease could open doors for metabolic interventions that specifically target these intercellular communication pathways.

3. Experimental Validation: The specific cell-cell and ligand-receptor pairs identified in this analysis warrant further experimental validation. Techniques such as spatial transcriptomics, multiplex immunohistochemistry, or *in vitro* co-culture systems could be used to confirm these interactions *in situ* and investigate their functional consequences in disease models. This would include detailed studies on the role of specific Acinar-Ductal interactions in regulating immune cell infiltration, beta cell protection, or fibrotic responses in T1D.

---

References:

1. FGF10 pancreatic development: PubMed search for "FGF10 pancreas development" https://pubmed.ncbi.nlm.nih.gov/?term=FGF10+pancreas+development
2. FGF10 regeneration: PubMed search for "FGF10 pancreatic regeneration" https://pubmed.ncbi.nlm.nih.gov/?term=FGF10+pancreatic+regeneration
3. ERBB4 GeneCards: GeneCards entry for ERBB4 https://www.genecards.org/cgi-bin/carddisp.pl?gene=ERBB4
4. NOTCH signaling pancreas: PubMed search for "NOTCH signaling pancreas development diabetes" https://pubmed.ncbi.nlm.nih.gov/?term=NOTCH+signaling+pancreas+development+diabetes
5. Integrin function: PubMed search for "integrin extracellular matrix cell adhesion pancreas" https://pubmed.ncbi.nlm.nih.gov/?term=integrin+extracellular+matrix+cell+adhesion+pancreas
6. Cholesterol signaling: PubMed search for "cholesterol signaling pancreatic inflammation" https://pubmed.ncbi.nlm.nih.gov/?term=cholesterol+signaling+pancreatic+inflammation
7. TGFB2 GeneCards: GeneCards entry for TGFB2 https://www.genecards.org/cgi-bin/carddisp.pl?gene=TGFB2

7. Macrophage Condition-Specific Surfaceome Markers in Pancreas

[Analysis Visualization Results]...

Analysis Overview

This analysis aimed to identify surfaceome-specific marker genes for Macrophages across different conditions (type1_diabetes, non_diabetic, autoantibody_positive) within the human pancreas, using single-cell RNA sequencing data. The plot_markers_and_expression_dot tool was employed to visualize differentially expressed genes (DEG) with specific criteria: only surfaceome markers were considered, up to 50 markers were sought per condition, with a minimum fold change of 1.5 and an adjusted p-value cutoff of 0.05. The dot plot displays the mean expression level (color intensity) and the fraction of cells expressing the gene (dot size) for each identified marker in the specified cell type and condition.

Visual Summary

The provided dot plot visualizes a subset of macrophage condition-specific markers.

• The plot highlights markers for the autoantibody_positive condition.
• Only two genes, TRHDE and LDLRAD3, are displayed.
• For both TRHDE and LDLRAD3 in autoantibody_positive macrophages (labeled "MM_403", likely referring to Macrophage), the dots are relatively small and light-colored. This indicates low mean expression levels (closer to the -0.1 to 0.0 range on the scaled expression scale) and a low fraction of cells expressing these genes within the macrophage population.
• A numerical indicator "45" is present adjacent to the "autoantibody_positive" condition, implying that the analysis identified 45 surfaceome-specific markers for this condition. However, only these two are shown in the current visualization, suggesting a highly truncated or partial display of the complete set of identified markers.

Biological Interpretation

The goal of identifying condition-specific surfaceome markers in macrophages is highly relevant in the context of pancreatic diseases, particularly Type 1 Diabetes (T1D) where autoantibody positivity often precedes clinical onset. Macrophages are key immune cells involved in inflammation, tissue remodeling, and immune regulation, and their specific phenotypes and functions can vary significantly across disease states.

• Macrophage Role in Pancreatic Disease: Macrophages are known to infiltrate pancreatic islets and contribute to islet inflammation (insulitis) and beta cell destruction in T1D. Understanding their condition-specific molecular signatures, especially surface markers, is crucial for unraveling disease mechanisms and identifying potential therapeutic targets.
• Significance of Surfaceome Markers: Surface proteins are particularly valuable as biomarkers because they are readily accessible for diagnostic detection (e.g., via flow cytometry) and can serve as targets for cell-specific therapeutic interventions (e.g., antibody-drug conjugates).
• Interpreting Visible Markers (TRHDE, LDLRAD3):
• TRHDE (Thyroid Hormone Degrading Enzyme): While primarily known for its role in thyroid hormone metabolism, deiodinases can have broader implications in immune cell function and inflammation. Macrophages are metabolically active cells, and altered expression of metabolic enzymes like TRHDE could reflect shifts in their metabolic state, potentially impacting their inflammatory or immune-regulatory functions in autoantibody-positive individuals. GeneCards: TRHDE
• LDLRAD3 (Low Density Lipoprotein Receptor Related Protein 3): This gene belongs to the LDL receptor family, which is involved in lipid uptake and signaling. Macrophages play a critical role in lipid metabolism, and dysregulation of LDL receptor pathways can influence macrophage polarization, foam cell formation, and inflammatory responses. Changes in LDLRAD3 expression could indicate altered lipid handling or signaling in macrophages within the autoantibody-positive state, potentially contributing to metabolic stress or inflammation in the pancreas. GeneCards: LDLRAD3
• Broader Set of Markers: Despite the low expression and prevalence of TRHDE and LDLRAD3 in the visual, the fact that 45 surfaceome markers were identified for autoantibody-positive macrophages suggests that there is a more extensive and potentially significant surfaceome remodeling occurring in these cells. These additional markers, if visible, would likely provide a richer picture of the macrophage phenotype in this disease-relevant condition.

Clinical or Translational Implications

The identification of condition-specific surfaceome markers for macrophages in autoantibody-positive individuals holds significant clinical and translational potential, even based on the broader analysis results (45 markers found) rather than just the two visible genes.

• Biomarker Development: The full set of 45 identified surfaceome markers could serve as a valuable panel for distinguishing macrophages in the autoantibody-positive state from those in non-diabetic or T1D conditions. This distinction could aid in:
• Early Disease Detection: Identifying individuals at high risk for T1D progression.
• Prognostic Assessment: Predicting disease trajectory or response to interventions.
• Therapeutic Targeting: Surfaceome markers are excellent candidates for targeted therapies. If certain markers are highly specific to pathogenic macrophages in the autoantibody-positive state, they could be explored as targets for immunomodulatory drugs, antibody therapies, or cell-specific gene delivery to modify macrophage activity and potentially halt or slow T1D progression.
• Further Validation: Given the limited visualization, further investigation of the full set of 45 surfaceome markers is warranted. This would involve examining their individual expression patterns, validating their specificity and functional roles, and assessing their utility in larger patient cohorts. Techniques such as flow cytometry, mass cytometry, or immunohistochemistry could be used for experimental validation of these surface markers.

8. Fibroblast Condition-Specific Surfaceome Markers in Type 1 Diabetes

[Analysis Visualization Results]...

Analysis Overview

This analysis aimed to identify condition-specific surfaceome markers in Fibroblast cells from pancreatic single-cell RNA-seq data. Specifically, it focused on identifying markers within the 'type1_diabetes' condition, comparing different samples or donors. The plot_markers_and_expression_dot tool was used, configured to find up to 50 surfaceome markers per condition with specific statistical cutoffs and to visualize their expression patterns.

Visual Summary

The dot plot visualizes the expression of four fibroblast-specific surfaceome markers: TMEM132C, TENM4, PCNX2, and NOX4, specifically within the 'type1_diabetes' condition. The y-axis represents two distinct sample groups (likely individual samples or donors) within this condition, labeled MM_401 and MM_380.

• Dot Size: Indicates the fraction of cells within each sample group that express a given gene.
• Dot Color Intensity: Represents the mean expression level of the gene in the expressing cells of that group (darker red signifies higher mean expression).
• The numbers '42' and '59' next to the bars on the right of the y-axis likely denote the number of Fibroblast cells identified in samples MM_401 and MM_380, respectively, that passed the minimum cell count threshold for marker discovery (N_cells_per_group_min=40).

A clear pattern emerges:

• In sample MM_401, all four genes (TMEM132C, TENM4, PCNX2, NOX4) show robust expression, indicated by relatively large, dark red dots. This suggests a high fraction of fibroblasts in MM_401 express these markers at considerable levels. Notably, TMEM132C and NOX4 exhibit the largest dot sizes and darkest colors, indicating broad expression across cells and high mean expression.
• In stark contrast, sample MM_380 shows minimal to no expression of these four genes, indicated by very small, light-colored or almost imperceptible dots. This suggests that these genes are either not expressed or expressed at very low levels in a negligible fraction of fibroblasts in sample MM_380.

Biological Interpretation

The differential expression of these surfaceome markers between sample MM_401 and MM_380, both from the 'type1_diabetes' condition, highlights significant heterogeneity in fibroblast activation or state within individuals affected by Type 1 Diabetes.

• Fibroblasts in Type 1 Diabetes: Fibroblasts are crucial stromal cells in the pancreas, contributing to extracellular matrix remodeling and tissue architecture. In T1D, fibroblasts can become activated, contributing to inflammation and fibrosis within the islets and surrounding pancreatic tissue, which may impact beta-cell survival and function.

Marker Genes and Pancreatic Fibroblast Biology

• TMEM132C (Transmembrane protein 132C): While its precise role in pancreatic fibroblasts and T1D is not extensively studied, transmembrane proteins are typically involved in cell-surface signaling, adhesion, or transport. Its differential expression suggests a unique cellular state or interaction capacity.
• TENM4 (Teneurin transmembrane protein 4): Teneurins are involved in cell adhesion and neural development, and emerging evidence suggests roles in diverse cellular processes, including cancer and organ development. Its presence on fibroblasts may indicate altered cell-cell or cell-matrix interactions.
• PCNX2 (Pecanex homolog 2): Pecanex proteins are known to interact with components of the Notch signaling pathway, which is a fundamental regulator of cell fate, differentiation, and tissue development. Notch signaling can influence fibroblast activation, proliferation, and matrix production, making PCNX2's expression potentially indicative of distinct fibroblast activation states.
• NOX4 (NADPH Oxidase 4): This gene encodes an enzyme that produces reactive oxygen species (ROS). NOX4 is a well-known contributor to oxidative stress, inflammation, and fibrosis in various tissues, including the pancreas and in the context of diabetes [GeneCards]. Elevated NOX4 expression in fibroblasts can drive fibrotic processes and exacerbate tissue damage. Its prominent expression in MM_401 suggests a highly activated, pro-fibrotic, or pro-inflammatory fibroblast phenotype in this particular type 1 diabetic sample.

The striking difference between samples MM_401 and MM_380 implies that not all individuals with T1D exhibit the same fibroblast responses. This could be due to variations in disease duration, severity, genetic background, or microenvironmental factors. The MM_401 sample's fibroblasts appear to be in a more activated or stress-responsive state, particularly given the high NOX4 expression, indicating increased oxidative stress or pro-fibrotic activity.

Clinical or Translational Implications

The identified surfaceome markers in pancreatic fibroblasts, particularly NOX4, hold significant translational potential:

• Biomarkers of Disease Heterogeneity: These genes could serve as biomarkers to stratify T1D patients based on their fibroblast activation state, potentially indicating varying degrees of pancreatic fibrosis or inflammation. This could inform personalized treatment approaches.
• Therapeutic Targets: NOX4 is a compelling therapeutic target. Given its role in oxidative stress and fibrosis, inhibiting NOX4 activity in activated fibroblasts could mitigate pancreatic damage and improve beta-cell function or survival in T1D [PubMed Search]. The other transmembrane proteins (TMEM132C, TENM4, PCNX2) could also represent novel targets if their functional roles in T1D-associated fibroblast pathology are elucidated.
• Experimental Validation: Further studies are warranted to validate the protein expression of these markers on pancreatic fibroblasts in T1D using techniques like immunohistochemistry or flow cytometry. Correlating marker expression with clinical parameters (e.g., C-peptide levels, autoantibody titers, disease duration) would strengthen their utility as prognostic or predictive tools. Isolation and functional characterization of these distinct fibroblast subsets could also shed light on their specific contributions to T1D pathophysiology.

9. Gene Ontology (GSA) Analysis of Upregulated Pathways in Pancreatic Cell Types Across Diabetic Conditions

[Analysis Visualization Results]...

Analysis Overview

This analysis presents Gene Ontology (GO) enrichment results, specifically using WikiPathways (WP) as the gene set database, for genes significantly upregulated (GSA_up) in specific pancreatic cell types (Acinar, Alpha, Beta, Ductal cells) across different conditions (autoantibody_positive, non_diabetic, type1_diabetes) compared to "others" (all other conditions combined, excluding the one being tested). The visualization is a dot plot where each dot's size and color intensity reflect the statistical significance (-log10(P-value)) of pathway enrichment.

Visual Summary

The dot plot effectively illustrates enriched pathways across 12 distinct comparison groups (4 cell types x 3 conditions/comparisons).

• X-axis: Represents the specific "cases," combining cell type and the condition being compared against "others" (e.g., "Acinar cell: autoantibody_positive_vs_others").
• Y-axis: Lists the WikiPathways terms, providing biological context for the enriched gene sets.
• Dot Size and Color: Both variables encode the -log10(P-value) of enrichment. Larger and darker red dots indicate more significant pathway enrichment.

Several pathways show strong and consistent enrichment across multiple cell types and conditions, while others appear to be more cell type- or condition-specific. Pathways related to inflammatory responses (e.g., IL-1, IL-2, IL-6, TNF alpha signaling) are particularly pronounced in Beta cells. Pathways associated with protein synthesis and processing (e.g., mRNA Processing, Translation Factors) are widely enriched across Acinar, Ductal, and Beta cells. Signaling pathways critical for cell growth and survival (e.g., Focal Adhesion-PI3K-Akt-mTOR, EGF/EGFR, MAPK) are also frequently observed, especially in Alpha and Beta cells.

Biological Interpretation

This GSA analysis reveals distinct and shared cellular perturbations across various pancreatic cell types in the context of Type 1 Diabetes (T1D) pathogenesis, characterized by autoantibody positivity and established T1D.

Cell Type-Specific and Shared Pathway Dysregulation:

1. Beta Cells: A Hub of Inflammatory and Stress Responses

• In both autoantibody_positive and type1_diabetes conditions, Beta cells show the most profound upregulation of inflammatory signaling pathways. Notably, TNF alpha Signaling Pathway (WP231), IL-1 Signaling Pathway (WP195), IL-2 Signaling Pathway (WP49), and IL-6 Signaling pathway (WP364) are highly enriched. This strong inflammatory signature is a hallmark of insulitis, the immune-mediated destruction of beta cells characteristic of T1D. Elevated cytokines like IL-1β, TNF-α, and IFN-γ are known to induce beta cell apoptosis and impair insulin secretion PubMed search: cytokines beta cell T1D apoptosis.
• Insulin Signaling Pathway (WP481) is also significantly enriched in beta cells, particularly in the type1_diabetes condition. While counterintuitive for T1D (where beta cells are destroyed), this could represent an initial compensatory response to maintain function or a dysregulation of the pathway contributing to beta cell stress and demise.
• Pathways related to protein synthesis and processing (mRNA Processing WP411, Translation Factors WP107) are also strongly upregulated in T1D beta cells, potentially indicating increased cellular stress, ER stress, or a maladaptive attempt to maintain protein homeostasis amidst inflammatory assault.
• Focal Adhesion-PI3K-Akt-mTOR-signaling pathway (WP3932) and MAPK Signaling Pathway (WP382) are enriched, suggesting altered cell survival, proliferation, or metabolic regulation that could be either protective or detrimental depending on the context of inflammation.

2. Alpha Cells: Metabolic and Growth Pathway Alterations

• Alpha cells in both autoantibody_positive and type1_diabetes conditions exhibit significant enrichment in Focal Adhesion-PI3K-Akt-mTOR-signaling pathway (WP3932), EGF/EGFR Signaling Pathway (WP437), and Cytoplasmic Ribosomal Proteins (WP477). These pathways are crucial for cell growth, proliferation, and metabolic regulation. Dysregulation here might reflect compensatory changes in alpha cell function (e.g., glucagon secretion) or an adaptive response to the changing pancreatic microenvironment in T1D GeneCards: EGF.
• Insulin Signaling Pathway (WP481) and MAPK Signaling Pathway (WP382) are also notably enriched in alpha cells in type1_diabetes, suggesting a broader metabolic reprogramming or an attempt to adapt to insulin deficiency.

3. Acinar and Ductal Cells: Evidence of Metabolic Reprogramming and Stress

• Both Acinar and Ductal cells in autoantibody_positive and type1_diabetes conditions show consistent upregulation of mRNA Processing (WP411) and Translation Factors (WP107). This suggests a general increase in protein synthesis machinery or a stress-induced upregulation of processes to handle misfolded proteins or adapt to metabolic demands within the pancreatic exocrine compartment.
• VEGFA-VEGFR2 Signaling Pathway (WP3888) is also enriched in these cell types, particularly in type1_diabetes. This pathway is critical for angiogenesis and vascular permeability UniProt: VEGFA. Its upregulation could indicate tissue remodeling, response to hypoxia, or inflammation-induced vascular changes in the diabetic pancreas.
• Focal Adhesion-PI3K-Akt-mTOR-signaling pathway (WP3932) is also enriched in these cell types in T1D/autoantibody positive states, indicating broader alterations in cell adhesion, migration, and growth pathways across the pancreatic tissue.

Pan-Pancreatic Observations:

• General Stress/Metabolic Shift: The widespread enrichment of mRNA Processing and Translation Factors across Acinar, Ductal, and Beta cells (and to a lesser extent, Alpha cells) in disease conditions points towards a general cellular stress response or altered metabolic state requiring increased protein synthesis.
• Growth/Survival Signaling: Pathways like Focal Adhesion-PI3K-Akt-mTOR and EGF/EGFR are broadly active, suggesting attempts at repair, regeneration, or altered survival mechanisms across various pancreatic cell types under diabetic stress.

Clinical or Translational Implications

The findings highlight several pathways with potential clinical relevance for Type 1 Diabetes:

• Targeting Beta Cell Inflammation: The robust inflammatory signature in beta cells (TNF-α, IL-1, IL-2, IL-6 signaling) strongly supports ongoing research into immune-modulating therapies to preserve beta cell mass and function in T1D. Specific inhibition of these cytokine pathways could be investigated as therapeutic strategies, particularly during the early, autoantibody-positive phase PubMed search: T1D immune therapy beta cell preservation.
• Understanding Pancreatic Remodeling: The upregulation of VEGFA-VEGFR2 Signaling in acinar and ductal cells suggests vascular changes or tissue remodeling within the pancreas. Modulating this pathway could potentially influence the microenvironment surrounding islets and impact disease progression.
• Monitoring Broader Pancreatic Health: Changes in protein synthesis and growth factor signaling across multiple cell types indicate that T1D pathology extends beyond just beta cells, affecting the broader pancreatic tissue. Biomarkers reflecting these pathways in acinar or ductal cells could offer insights into overall pancreatic health and disease severity.
• Insulin Signaling in Alpha and Beta Cells: The complex enrichment of Insulin Signaling pathways in both alpha and beta cells warrants further investigation. This could represent compensatory mechanisms or dysregulation contributing to both hyperglycemia and beta cell failure in T1D.

10. Gene Set Enrichment Analysis of Pancreatic Cell Types Across Diabetes Conditions

[Analysis Visualization Results]...

Analysis Overview

This analysis presents Gene Set Enrichment Analysis (GSEA) results for seven key pancreatic cell types (Acinar cell, Alpha cell, Beta cell, Ductal cell, Fibroblast, Macrophage, and Stellate cell) across three conditions: 'type1_diabetes', 'non_diabetic', and 'autoantibody_positive'. Each condition is compared against "others" (i.e., all other conditions combined) to identify significantly enriched or depleted pathways. The dot plot visualizes the Normalized Enrichment Score (NES) using a divergent colormap (RdBu_r), where red indicates enrichment (positive NES, upregulated pathways) and blue indicates depletion (negative NES, downregulated pathways). The size of each dot corresponds to the statistical significance (-log(p-value)), with larger dots indicating higher significance.

Visual Summary

The dot plot effectively displays the most significant pathway enrichments and depletions across various cell types and conditions.

• Widespread Inflammatory and Stress Responses in Disease: A prominent pattern is the consistent enrichment (red dots) of inflammatory pathways, such as "Inflammatory Response Pathway WP453" and "IL-1 signaling pathway WP195," across most cell types (Acinar, Alpha, Beta, Ductal, Macrophage) in both 'autoantibody_positive_vs_others' and 'type1_diabetes_vs_others' comparisons. "DNA Damage Response (only ATM dependent) WP710" also shows frequent enrichment in these disease states. This suggests a systemic cellular stress and immune activation within the pancreas during the progression and presence of Type 1 Diabetes (T1D).
• Beta Cell Specific Responses in T1D: Beta cells, central to T1D pathology, show strong enrichment of "Inflammatory Response Pathway WP453," "IL-1 signaling pathway WP195," and "Toll-like Receptor Signaling Pathway WP75" in 'type1_diabetes_vs_others'. Interestingly, "Type II diabetes mellitus WP1584" is also enriched in Beta cells under T1D, suggesting shared stress or dysfunction pathways between diabetes types, or that the pathway broadly captures beta cell pathology. Conversely, "Vitamin B12 Metabolism WP1533" and "One Carbon Metabolism WP241" pathways appear depleted (blue dots) in Beta cells in 'type1_diabetes_vs_others', indicating metabolic dysregulation.
• Stromal Cell Remodeling in T1D: Fibroblasts and Stellate cells, key stromal components, exhibit enrichment of pathways related to extracellular matrix (ECM) remodeling, such as "Matrix Metalloproteinases WP129," "miRNA targets in ECM and membrane receptors WP2911," and "TGF-beta Signaling Pathway WP366" in 'type1_diabetes_vs_others'. This is indicative of fibrotic processes and tissue restructuring within the pancreas in T1D.
• Macrophage Activation: Macrophages, as expected, show substantial enrichment of immune-related pathways (e.g., "Inflammatory Response Pathway WP453," "IL-1 signaling pathway WP195," "Toll-like Receptor Signaling Pathway WP75") in 'autoantibody_positive_vs_others' and 'type1_diabetes_vs_others', highlighting their active role in the immune surveillance and inflammatory milieu of the pancreas.
• Non-Diabetic Controls: The 'non_diabetic_vs_others' comparisons generally show fewer highly enriched pathways (fewer large, intensely colored dots) or, in some cases, depletion of pathways that are enriched in disease states. This provides a baseline contrasting the disease conditions.

Biological Interpretation

The GSEA results provide clear insights into the cell-type-specific and pan-cellular biological processes disrupted in the context of autoantibody positivity and Type 1 Diabetes in the pancreas.

• Broad Inflammatory Landscape: The widespread activation of "Inflammatory Response Pathway" and "IL-1 signaling pathway" across multiple pancreatic cell types (Acinar, Alpha, Beta, Ductal, Macrophage) in autoantibody-positive and T1D conditions points to a chronic inflammatory environment. IL-1β, a potent pro-inflammatory cytokine, is known to be elevated in the islets of T1D patients and contributes to beta cell dysfunction and death. Activation of "Toll-like Receptor Signaling Pathway" in Beta cells and Macrophages further supports an innate immune response, potentially triggered by cellular stress or pathogen-associated molecular patterns (PAMPs)/damage-associated molecular patterns (DAMPs) released during early autoimmunity or beta cell demise [1].
• Beta Cell Vulnerability and Metabolic Reprogramming: The strong inflammatory and stress signals in Beta cells in T1D highlight their direct involvement and vulnerability. The enrichment of "DNA Damage Response" pathways suggests increased cellular stress beyond inflammation. The depletion of "Vitamin B12 Metabolism" and "One Carbon Metabolism" pathways indicates metabolic dysregulation in beta cells, which could impair insulin synthesis and secretion or alter their epigenetic landscape, contributing to dysfunction [2]. The enrichment of "Type II diabetes mellitus" pathway in T1D beta cells could reflect common pathways of beta cell failure or stress, rather than specific T2D etiology.
• Role of Stromal Cells in Pancreatic Remodeling: The activation of matrix remodeling pathways (e.g., "Matrix Metalloproteinases," "TGF-beta Signaling Pathway") in Fibroblasts and Stellate cells is consistent with the development of pancreatic fibrosis, a common feature observed in T1D pancreases. This remodeling can affect islet architecture, blood supply, and communication between pancreatic cells, potentially hindering beta cell regeneration or survival [3].
• Immune Cell Infiltration and Activation: The strong inflammatory signatures in Macrophages reinforce their crucial role in immune responses within the pancreas during T1D progression. Activated macrophages can contribute to islet inflammation and beta cell destruction through cytokine release and antigen presentation [4].

Clinical or Translational Implications

The findings have several clinical and translational implications:

• Biomarker Discovery: The identified cell-type-specific pathway enrichments, particularly in autoantibody-positive individuals, could yield novel biomarkers for early disease detection or risk stratification. For example, gene signatures related to inflammatory pathways in Acinar or Alpha cells might precede overt T1D.
• Therapeutic Targets: Targeting the consistently upregulated inflammatory pathways (e.g., IL-1 signaling, Toll-like receptor signaling) across multiple cell types represents a potential therapeutic strategy to mitigate immune-mediated destruction in T1D. Anti-inflammatory agents could be beneficial, especially in the early stages of the disease.
• Addressing Fibrosis: Modulating ECM remodeling and fibrotic pathways in Fibroblasts and Stellate cells could be a strategy to preserve pancreatic architecture and function, potentially improving beta cell survival or facilitating regenerative approaches.
• Metabolic Interventions: The observed metabolic dysregulation in Beta cells (e.g., Vitamin B12 and One Carbon Metabolism) suggests potential for nutritional or metabolic interventions to support beta cell health in T1D.
• Multi-Cellular Therapeutic Approaches: The involvement of diverse cell types in the disease process underscores the need for multi-pronged therapeutic strategies that consider not only immune cells and beta cells but also stromal and exocrine components of the pancreas.

---

References

1. IL-1 signaling in T1D: Maedler, K., et al. (2002). Glucose-induced beta-cell production of IL-1beta contributes to glucotoxicity in human pancreatic islets. *Journal of Clinical Investigation, 110*(6), 851–861. PubMed Search: IL-1 beta type 1 diabetes beta cell
2. One Carbon Metabolism & Beta Cells: Pan, C., et al. (2020). One-Carbon Metabolism: A Linker Between Nutrition and Beta-Cell Function. *Frontiers in Endocrinology, 11*, 571437. PubMed Search: one carbon metabolism beta cell diabetes
3. Pancreatic Fibrosis in T1D: Rix, N., et al. (2018). Pancreatic fibrosis in Type 1 diabetes. *Diabetologia, 61*(6), 1438–1448. PubMed Search: pancreatic fibrosis type 1 diabetes
4. Macrophage Role in T1D: Ferris, S. T., & Gannon, M. (2021). Pancreatic Macrophages in Type 1 Diabetes: Initiators, Perpetrators, or Protective Pods? *Frontiers in Endocrinology, 12*, 674987. PubMed Search: macrophage type 1 diabetes pathogenesis

11. Discussion

The single-cell RNA-sequencing analysis of human pancreas from non-diabetic, autoantibody-positive, and type 1 diabetes (T1D) individuals reveals a complex interplay of cellular and molecular perturbations underlying T1D pathogenesis. A hallmark finding is the profound reduction or near absence of Beta cells in many T1D samples, directly confirming the immune-mediated destruction of insulin-producing cells, while Alpha cells appear relatively more resilient. Individuals with autoantibody positivity show Beta cell proportions similar to non-diabetic controls, suggesting that significant population-level loss may occur later in disease progression or is highly heterogeneous in this pre-diabetic stage.

Immune cell populations exhibit striking re-distributions. Macrophages in T1D samples show a clear and consistent dominance of the pro-inflammatory M1 phenotype, often exceeding 80-90% of the total macrophage pool. In contrast, non-diabetic controls maintain a more diverse macrophage composition, with a notable presence of immunomodulatory M2C macrophages. The autoantibody-positive group displays an intermediate, heterogeneous macrophage profile, with some samples showing an increase in M1 polarization, indicating an evolving pro-inflammatory environment even before clinical T1D onset. This M1 dominance strongly implicates these cells in driving insulitis and beta cell destruction.

Cell-cell interaction analysis highlights critical communication axes between pancreatic Acinar and Ductal cells. Conserved growth factor signaling (EGF-EGFR, FGF10-FGFR2, NRG-ERBB, JAG1-NOTCH) and cell adhesion pathways (NECTIN/CADM) are observed across all conditions, suggesting their fundamental roles in pancreatic homeostasis. However, condition-specific shifts are evident; for instance, the LAMC1-integrin α6β1 complex interaction becomes more prominent in Acinar-Acinar and Acinar-Ductal cells in T1D, pointing to potential extracellular matrix remodeling or altered cell adhesion that could influence tissue integrity and immune cell infiltration. Altered cholesterol and glutamate-mediated interactions also suggest changes in metabolic communication in disease.

Condition-specific marker analysis, though partially visualized for macrophages, revealed 45 surfaceome markers for autoantibody-positive macrophages, including TRHDE and LDLRAD3, which could reflect altered metabolic or signaling states. For fibroblasts in T1D, NOX4 emerged as a key surfaceome marker, showing robust expression in some T1D samples (e.g., MM_401) but not others, highlighting patient heterogeneity in fibroblast activation. NOX4's role in oxidative stress and fibrosis suggests a pro-fibrotic or pro-inflammatory fibroblast phenotype in affected individuals.

Gene Ontology (GSA) and Gene Set Enrichment Analysis (GSEA) provide a comprehensive view of pathway dysregulation. Beta cells in both autoantibody-positive and T1D conditions show profound upregulation of inflammatory pathways, including TNF-α, IL-1, IL-2, IL-6, and Toll-like Receptor signaling, underscoring the central role of chronic inflammation in beta cell destruction. Beyond inflammation, Beta cells in T1D also exhibit metabolic dysregulation, with a depletion of Vitamin B12 and One Carbon Metabolism pathways, which could impair insulin synthesis and secretion or alter epigenetic landscapes. Alpha cells show alterations in metabolic and growth pathways (e.g., PI3K-Akt-mTOR, EGF/EGFR signaling), potentially reflecting compensatory changes. Acinar and Ductal cells also demonstrate evidence of cellular stress (mRNA processing, translation factors) and vascular remodeling (VEGFA-VEGFR2 signaling) in disease conditions, indicating that T1D pathology extends beyond the islets to affect the broader pancreatic tissue. Importantly, stromal cells like Fibroblasts and Stellate cells in T1D exhibit activated matrix remodeling pathways (Matrix Metalloproteinases, TGF-beta Signaling), consistent with the development of pancreatic fibrosis.

Hypotheses:

1. The prominent shift to M1-dominant macrophages in Type 1 Diabetes directly drives beta cell apoptosis and impairs insulin secretion via specific pro-inflammatory cytokine pathways.
2. Altered LAMC1-integrin α6β1 complex interactions between acinar and ductal cells in Type 1 Diabetes contribute to extracellular matrix remodeling and pancreatic fibrosis, exacerbating islet dysfunction.
3. Dysregulation of Vitamin B12 and One Carbon Metabolism in beta cells compromises their metabolic resilience and increases their susceptibility to immune-mediated destruction in Type 1 Diabetes.
4. Fibroblast activation, specifically marked by elevated NOX4 expression, promotes an oxidative stress and pro-fibrotic microenvironment that contributes to beta cell damage and impaired pancreatic function in Type 1 Diabetes.

Potential therapeutic targets:

1. IL-1 and TNF-alpha signaling pathways: These pro-inflammatory cytokine pathways are strongly upregulated in Beta cells and Macrophages in T1D, driving insulitis and beta cell destruction. Evidence: GSA/GSEA analyses show significant enrichment of IL-1 and TNF-alpha signaling pathways in Beta cells and Macrophages in both autoantibody-positive and type 1 diabetes conditions. Validation: Test anti-IL-1β or anti-TNF-α therapies in preclinical models of T1D to assess beta cell preservation and disease progression. Clinical trials of anti-cytokine therapies in new-onset or high-risk T1D patients.
2. NOX4 (NADPH Oxidase 4): NOX4 is implicated in oxidative stress and fibrotic processes. Its increased expression in pancreatic fibroblasts from T1D patients suggests it contributes to tissue damage and remodeling. Evidence: Condition-specific marker analysis shows prominent expression of NOX4 in fibroblasts from certain type 1 diabetes samples (e.g., MM_401), and NOX4 is a known driver of fibrosis in other contexts. Validation: Evaluate the efficacy of small molecule NOX4 inhibitors in *in vivo* models of pancreatic fibrosis and T1D. Conduct *in vitro* studies to confirm NOX4's role in fibroblast activation and ECM production in response to diabetic stressors.
3. M1 Macrophage Polarization Pathways: The pancreatic macrophage population shifts dramatically towards a pro-inflammatory M1 phenotype in T1D, actively contributing to beta cell damage. Modulating this polarization could mitigate disease. Evidence: Population bar plots demonstrate a striking dominance of M1 macrophages in type 1 diabetes samples. GSEA shows enrichment of Toll-like Receptor Signaling pathways in macrophages, which are known to drive M1 polarization. Validation: Investigate small molecule inhibitors or genetic approaches that specifically inhibit M1 macrophage polarization or promote M2 differentiation in T1D models. Assess changes in pancreatic immune cell composition and beta cell survival.
4. Integrin α6β1 complex: Increased LAMC1-integrin α6β1 interaction in T1D suggests altered cell adhesion and extracellular matrix remodeling, potentially impacting islet architecture and contributing to fibrosis. Evidence: Cell-cell interaction analysis shows increased prominence of LAMC1_integrin_a6b1_complex in Acinar-Acinar and Acinar-Ductal cell interactions in type 1 diabetes. Validation: Test the effect of integrin α6β1 antagonists in pancreatic organoid models or *in vivo* T1D models to determine their impact on pancreatic fibrosis, cell adhesion, and beta cell function.

Follow-up validation ideas:

1. To validate the role of M1 macrophages, utilize flow cytometry or spatial transcriptomics to quantify M1 macrophage infiltration and proximity to beta cells in T1D pancreatic tissue. Functional *in vitro* co-culture experiments can then assess the impact of M1-polarized macrophages on beta cell viability and function using cytokine neutralization or genetic knockdown approaches.
2. To confirm altered LAMC1-integrin interactions, perform multiplex immunohistochemistry or immunofluorescence on T1D pancreatic sections to visualize LAMC1 and integrin α6β1 co-localization and protein levels. *In vitro* perturbation assays using integrin-blocking antibodies or genetic manipulation in pancreatic fibroblast and acinar/ductal co-cultures can elucidate functional consequences on adhesion, migration, and ECM production.
3. For metabolic dysregulation in beta cells, conduct targeted metabolomics on isolated primary human T1D beta cells to quantify intermediates of Vitamin B12 and One Carbon Metabolism. Genetic knockdown or overexpression of key enzymes in these pathways in human beta cell lines, followed by functional assays under inflammatory stress, can confirm their impact on beta cell survival and insulin secretion.
4. To validate fibroblast activation and NOX4 involvement, perform immunohistochemistry for NOX4 in pancreatic biopsies from T1D patients, correlating expression with fibrosis markers (e.g., collagen) and clinical parameters. *In vitro* studies with primary human pancreatic fibroblasts exposed to T1D-relevant inflammatory cytokines can induce NOX4 expression, allowing testing of specific NOX4 inhibitors on ROS production and fibrogenic gene expression.
5. For the macrophage surfaceome markers TRHDE and LDLRAD3, flow cytometry on immune cells isolated from autoantibody-positive individuals can confirm their surface expression and identify specific macrophage subsets. Functional studies, such as gene knockdown in macrophages, could then investigate the role of these markers in macrophage activation, migration, or interaction with other pancreatic cells.

Limitations:

This report is based on a cross-sectional single-cell RNA-seq dataset, which limits the ability to infer causality or track disease progression over time. The analysis primarily focuses on pancreatic cells and interactions, and therefore may not fully capture systemic immune responses contributing to Type 1 Diabetes. While cell type annotations are robust, the presence of 'unassigned' cells indicates potential for further characterization of rare or novel populations. Furthermore, the 'surfaceome_only' marker discovery, while valuable, occasionally included non-surface proteins if they were canonical markers, which requires careful interpretation for therapeutic targeting. The term 'others' in GSA/GSEA comparisons pools different conditions, necessitating cautious interpretation of relative enrichment/depletion. Translational implications drawn from this analysis require extensive experimental validation in *in vitro* models, animal models, and ultimately, human clinical studies.

12. Query List

1. Show UMAP including condition, sample, major celltype, minor celltype, and celltype_subset in 2 columns and save.
2. Show major celltype scores on UMAP and save.
3. Show the overall Celltype_subset marker expression dot plot and save it. Keep target_cell as None and keep the other arguments at their default values.
4. Show population bar plot for minor cell types and save.
5. Show subset population bar plot for macrophages and save.
6. Show cell-cell interactions by condition and save the result. Select at most 80 cell-cell interactions for each condition.
7. Extract condition-specific markers for macrophages and show a dot plot. Select only surfaceome markers, up to 50 per condition, and save.
8. Extract condition-specific markers for fibroblasts and show a dot plot. Select only surfaceome markers, up to 50 per condition, and save.
9. Show Gene Ontology (GSA) analysis results bar plot for Acinar cell, Alpha cell, Beta cell, and Ductal cell, and save.
10. Show Gene set enrichment analysis results dot plot for Acinar cell, Alpha cell, Beta cell, Ductal cell, Fibroblast, Macrophage, and Stellate cell, using RdBu_r color map and n_pws_to_show = 80, and save.