SCODiA Report by MLBI Lab

Single-Cell Transcriptomic and Genomic Analysis Reveals Distinct Malignant Ductal Cell States and an Immunosuppressive Tumor Microenvironment in Pancreatic Ductal Adenocarcinoma

This comprehensive single-cell analysis of pancreatic ductal adenocarcinoma (PDAC) elucidates profound molecular and cellular shifts compared to adjacent normal pancreas. We identify malignant ductal cells by their prominent aneuploidy, extensive copy number variations, and dysregulated cell cycle. The tumor microenvironment exhibits significant remodeling, characterized by an increased abundance of pro-tumorigenic macrophages and stromal cells, alongside a shift in T cell populations towards immunosuppressive phenotypes. These findings highlight critical pathways and cellular interactions that drive PDAC progression and foster immune evasion.

Contents

0. Dataset overview
1. UMAP Visualization of Pancreatic Single-Cell RNA-Seq Data by Condition, Sample, Cell Type, and Ploidy
2. Major Cell Type Score and Annotation Overview on UMAP
3. Marker Expression Profile for Pancreatic Cell Subsets
4. Ductal Cell Copy Number Variation (CNV) Analysis in Pancreatic Cancer
5. UMAP Visualization of CNV Patterns Across Pancreatic Cell Types and Conditions
6. Cell Type Population Analysis of Minor Cell Types in Pancreatic Samples
7. T cell Subtype Population Analysis in Pancreatic Ductal Adenocarcinoma (PDAC)
8. Differential T cell Subset Proportions in Pancreatic Ductal Adenocarcinoma (PDAC)
9. Ductal Cell Ploidy Analysis in Pancreatic Adenocarcinoma (PDAC)
10. Pancreatic Ductal Adenocarcinoma (PDAC) Cell-Cell Interaction Landscape: Shifts Towards Immunosuppression and Pro-Tumorigenic Signaling
11. Pancreatic Cell-Cell Interaction Analysis of Immune Checkpoint and Cell Cycle Pathways in Normal and PDAC Conditions
12. Condition-Specific Cell-Cell Interaction Patterns in PDAC
13. Ductal Cell Condition-Specific Surfaceome Markers in Pancreatic Ductal Adenocarcinoma (PDAC)
14. Macrophage Condition-Specific Surfaceome Markers in Pancreatic Cancer
15. Comprehensive Surfaceome Marker Expression for Pancreatic Cell Subsets
16. T cell CD4+ Condition-Specific Surfaceome Markers in Pancreatic Cancer
17. Dysregulation of Cell Cycle Pathway Genes in Pancreatic Ductal Adenocarcinoma (PDAC) Ductal Cells
18. Ductal Cell Gene Ontology Analysis: Insights into Pancreatic Homeostasis and Pancreatic Ductal Adenocarcinoma (PDAC) Pathobiology
19. 췌장암(PDAC) 미세환경 내 세포 유형별 유전자 세트 농축 분석 (GSEA)
20. Discussion
21. Query List

0. Dataset overview

Dataset Summary

• Type: Single-cell RNA-seq data processed by SCODA.
• Dimensions: Contains 49,221 cells and 23,910 genes.
• Species & Tissue: Data is from human Pancreas tissue.
• Conditions: The samples are categorized into 'Adj_normal' and 'PDAC' conditions.
• Cell Type Annotations: Cells are hierarchically classified into major, minor, and subset levels, including types like Stromal cell, Endothelial cell, Acinar cell, T cell, Myeloid cell, Mast cell, Ductal cell, and B cell.
• Tumor Origin Celltype: Ductal cell is identified as the tumor origin cell type.
• Ploidy Information: Cells are categorized by 'ploidy_dec' as Aneuploid or Diploid.
• Reference Condition: 'Adj_normal' is the reference condition for analyses like DEG_vs_ref, GSEA_vs_ref, and GSA_vs_ref_up.

Precomputed Results

• Cell-Cell Interaction (CCI): Precomputed results are available at both condition-level (uns['CCI']) and sample-level (uns['CCI_sample']).
• Differential Expression Genes (DEG): DEG results (log2_FC, pval_adj, etc.) are precomputed for each 'celltype_minor' comparing one condition against the rest, stored in uns['DEG'].
• Gene Set Enrichment Analysis (GSEA): GSEA results are precomputed for each 'celltype_minor' comparing one condition against the rest, stored in uns['GSEA'].
• Gene Ontology (GO/GSA): GO (GSA) results for enriched pathways are precomputed for each 'celltype_minor' comparing one condition against the rest, stored in uns['GSA_up'].
• Copy Number Variation (CNV): CNV estimates (obsm['X_cnv']) and ploidy inference labels (obs['ploidy_dec']) are available.

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

[Analysis Visualization Results]...

Analysis Overview

This analysis presents UMAP (Uniform Manifold Approximation and Projection) plots derived from single-cell RNA-seq data of the human pancreas, comprising both adjacent normal (Adj_normal) and pancreatic ductal adenocarcinoma (PDAC) samples. The UMAPs visualize the cellular landscape, colored by various metadata features including disease condition, individual samples, major and minor cell types, cell type subsets, and ploidy status. These visualizations are critical for assessing data quality, understanding cellular heterogeneity, and evaluating the distinctness of different cell populations and disease states within the dataset.

Visual Summary

• condition UMAP: The UMAP clearly segregates cells based on condition. A large proportion of cells, particularly those forming the dense central clusters and several peripheral clusters, are colored purple, indicating they originate from PDAC samples. Adjacent normal cells (red) are largely distinct, forming their own separate clusters or occupying specific, well-delineated regions within larger clusters. This suggests a strong transcriptomic difference between cells from PDAC and adjacent normal tissues.
• sample UMAP: When colored by individual samples, the UMAP shows a good intermixing of cells from different samples, especially within the larger, central cell clusters. This indicates that batch effects are largely mitigated, and the underlying biological signals are preserved across samples. Notably, the adjacent normal samples (AdjN_1, AdjN_2, AdjN_3) tend to cluster together distinctly, reflecting their shared non-diseased state.
• celltype_major UMAP: Major cell types form well-defined, spatially separated clusters on the UMAP, demonstrating robust cell type annotation. For instance, Ductal cells (orange) occupy a prominent central region, T cells (light blue) form several distinct immune-related clusters, Stromal cells (teal/green) cluster separately, and Acinar cells (dark red) are primarily found in a distinct peripheral cluster. This indicates clear transcriptomic identities for these major cell lineages.
• celltype_minor UMAP: This plot provides a finer resolution of cell types. The major cell type clusters are further subdivided into minor cell populations. For example, T cells are resolved into T cell CD4+ and T cell CD8+, Macrophages (Mac) appear as a distinct cluster, and Stellate cells are clearly visible within the stromal compartment. The distinct separation of these minor cell types suggests high-quality and granular annotation.
• ploidy_dec UMAP: The ploidy status reveals a striking pattern. A substantial, contiguous cell cluster, largely overlapping with the main Ductal cell population, is predominantly marked as Aneuploid (dark red). The vast majority of other cell types are classified as Diploid (yellow). A small fraction of cells are labeled as 'Unclear'. This clear partitioning strongly suggests the identification of malignant (aneuploid) cells distinct from non-malignant (diploid) cells.
• celltype_subset UMAP: This UMAP displays the highest level of cell type granularity. It further refines the minor cell types into specific subsets, such as various macrophage polarizations (Mac_M1, Mac_M2A-D), T cell helper subsets (Th1, Th2, Th9, Th17, Th22, Tfh), cytotoxic T cells (T_Cyto), naive T cells (T_Naive), regulatory T cells (Treg), and different B cell populations (B cell (Memory), BMZ, Bf, Breg). These subsets form distinct, albeit sometimes intermingled, patterns within their broader cell type clusters, providing a detailed view of cellular heterogeneity.

Biological Interpretation

The UMAP visualizations provide a comprehensive overview of the cellular landscape of the human pancreas in both adjacent normal and PDAC conditions.

• Disease-Specific Cellular Changes: The clear separation of PDAC cells from adjacent normal cells on the condition UMAP underscores the profound transcriptomic reprogramming that occurs in pancreatic cancer. This indicates that the disease state is a primary driver of cellular gene expression profiles, enabling robust differentiation between healthy and diseased cells.
• Cell Type Identity and Heterogeneity: The distinct clustering patterns observed across celltype_major, celltype_minor, and celltype_subset UMAPs confirm the successful identification and annotation of diverse cell populations within the pancreas. The hierarchical resolution, from broad major types to highly specific subsets (e.g., various T cell and macrophage subtypes), is crucial for dissecting the complex cellular ecosystem of the tumor microenvironment (TME) and the normal pancreas.
• Identification of Malignant Cells: The ploidy_dec UMAP strongly points to the identification of the malignant cell population. Given that PDAC originates from Ductal cells and aneuploidy is a hallmark of cancer, the large cluster of aneuploid cells predominantly co-localizing with Ductal cells (as seen in the celltype_major and celltype_minor plots) is highly indicative of the tumor cells. This separation is vital for downstream analyses focusing on cancer-specific alterations.
• Robust Data Integration: The general mixing of cells from different samples in the sample UMAP, while maintaining distinct condition and cell type clusters, suggests effective batch effect correction. This enhances confidence that observed differences are biological rather than technical artifacts, making the dataset reliable for further in-depth analyses of PDAC biology.
• Pancreatic Ductal Adenocarcinoma (PDAC) Context: The dominance of Ductal cells in the central cluster, combined with their aneuploid status, aligns perfectly with the understanding that PDAC is an aggressive cancer originating from the ductal epithelium of the pancreas. The presence and diverse subsets of immune cells (T cells, B cells, Macrophages) and stromal cells (Stellate, Fibroblast) within the tumor environment further highlight the complex interplay characteristic of the PDAC TME.

Annotation Notes

The UMAPs demonstrate high quality in cell type annotation and ploidy inference. Cells belonging to the same annotated group generally cluster together, and distinct groups are well separated, which is ideal for downstream differential expression, pathway analysis, and cell-cell interaction studies. The consistent clustering of adjacent normal samples also validates the reference group used in differential analyses. The 'Unclear' category in ploidy_dec could represent cells with ambiguous ploidy signals or those in a transition state, and their relatively small proportion indicates good confidence in the main aneuploid/diploid calls.

2. Major Cell Type Score and Annotation Overview on UMAP

[Analysis Visualization Results]...

Analysis Overview

This analysis presents a series of Uniform Manifold Approximation and Projection (UMAP) plots, visualizing single-cell RNA-sequencing data from pancreatic tissue (Adj_normal and PDAC conditions). The primary objective is to evaluate the distribution and quality of major cell type assignments (celltype_major) based on HiCAT scores, alongside an assessment of cellular ploidy (ploidy_dec) within the same embedding space. Each major cell type is represented by a score (HiCAT_major_score), indicating the strength of its characteristic gene expression signature for each cell. This provides an essential foundation for understanding the cellular landscape and validating cell type annotations in the dataset.

Visual Summary

The UMAP visualizations clearly delineate distinct cellular populations based on their transcriptional profiles.

• HiCAT_major_score Plots: Each plot highlights cells with high scores (yellow/green) for a specific major cell type.
• Immune cells (T cell, B cell, Myeloid cell, Mast cell) generally form well-separated clusters with high scores in distinct regions of the UMAP, suggesting robust and distinct transcriptional identities. For instance, T cells primarily populate a large cluster in the upper left, while Myeloid cells are concentrated in a cluster in the middle-left.
• Stromal cells and Endothelial cells also exhibit clear, localized high-score regions, indicating their unique transcriptional programs. Stromal cells occupy a broad region in the middle-left/center, and Endothelial cells form smaller, distinct clusters.
• Pancreatic exocrine (Acinar cell) and endocrine cells (Alpha, Beta, Delta, Epsilon, Gamma (PP) cell) show specific, often smaller, high-score regions. Acinar cells are prominent in the lower-right quadrant. The various islet cell types (Alpha, Beta, Delta, Epsilon, Gamma) show more compact, sometimes overlapping, distributions, consistent with their organization within pancreatic islets.
• Ductal cells, identified as the tumor origin cell type, show a large, prominent cluster with high scores located centrally on the UMAP, spanning a significant portion of the embedding. This suggests a substantial population of cells with a ductal phenotype.
• Other cell types such as Pancreatic progenitor cells and Schwann cells also display distinct, albeit smaller, high-score regions.
• Ploidy Status (ploidy_dec): The UMAP colored by ploidy_dec reveals a striking pattern. A large cluster of aneuploid cells (maroon) is highly localized and largely overlaps with the central, prominent Ductal cell cluster identified by both its HiCAT score and the celltype_major annotation. The majority of other cell types appear to be diploid (light yellow). Cells with 'Unclear' ploidy are also present, scattered across the map.
• Major Cell Type Annotation (celltype_major): The final celltype_major plot shows the assigned identities, which visually align very well with the high-score regions observed in the individual HiCAT_major_score plots. The distinct clustering and non-overlapping nature of most major cell types confirm the quality of the automated annotation. The large central cluster annotated as "Ductal cell" is particularly noticeable.

Biological Interpretation

The UMAP plots provide a clear cellular landscape of the human pancreas in the context of pancreatic ductal adenocarcinoma (PDAC).

1. High-Quality Cell Type Annotation: The strong correspondence between the HiCAT_major_score for each cell type and the final celltype_major assignments on the UMAP embedding indicates robust and reliable cell type identification. Each major cell type forms transcriptionally distinct clusters, validating the initial unsupervised clustering and subsequent annotation. This is crucial for downstream analyses, ensuring that comparisons and interpretations are based on accurately defined cell populations.
2. Identification of Putative Malignant Ductal Cells: A key observation is the substantial overlap between the cluster of cells highly scored as "Ductal cell" and the region predominantly labeled as "Aneuploid" in the ploidy_dec plot. Given that "Ductal cell" is identified as the "Tumor origin celltype" and aneuploidy is a hallmark of cancer, this strong co-localization strongly suggests that this cluster represents the malignant tumor cell population within the PDAC samples. These cells likely exhibit genomic instability and altered gene expression characteristic of pancreatic cancer.

• Reference for aneuploidy in cancer: PubMed search for "aneuploidy cancer hallmark"

3. Tumor Microenvironment Composition: The UMAP clearly shows the presence and distinct localization of various immune cells (T cells, B cells, Myeloid cells, Mast cells) and stromal cells (e.g., Stromal cell, Endothelial cell). These populations form separate transcriptional entities from the putative malignant ductal cells. Their spatial arrangement relative to the aneuploid ductal cell cluster (which appears to be centrally located, with other cell types often surrounding or interspersed) provides an initial visual indication of the complexity and cellular heterogeneity of the pancreatic tumor microenvironment (TME). Understanding the specific composition and distribution of these support cells is critical for comprehending tumor progression and therapeutic responses.

• Reference for pancreatic tumor microenvironment: PubMed search for "pancreatic tumor microenvironment"

4. Normal Pancreatic Cell Representation: The presence of distinct clusters for Acinar cells and various pancreatic islet cells (Alpha, Beta, Delta, Epsilon, Gamma) indicates that normal pancreatic tissue components are also well-represented and distinguishable within the dataset. This allows for comparative analyses between tumor-associated and normal tissue components, providing a baseline for identifying disease-specific changes.

Annotation Notes

The comprehensive display of major cell type scores, alongside ploidy and final annotations, confirms the high quality and specificity of the cell type assignments in this dataset. The strong visual concordance between individual cell type scores and the clustered annotations on the UMAP reinforces confidence in the cell identity labels. The clear segregation of cell types, and especially the distinct cluster of aneuploid ductal cells, provides a robust foundation for more detailed downstream analyses, such as differential gene expression, pathway analysis, and cell-cell interaction studies, with well-defined cell populations.

3. Marker Expression Profile for Pancreatic Cell Subsets

[Analysis Visualization Results]...

Analysis Overview

This analysis presents a dot plot illustrating the expression patterns of key marker genes across various celltype_subset populations identified in the single-cell RNA-seq data from human pancreatic tissue. The purpose of this visualization is to validate the distinct molecular identities of the annotated cell subsets by confirming the expression of known lineage-specific markers. The plot shows the mean expression level (color intensity) and the percentage of cells expressing a given gene (dot size) for each cell subset. Markers were selected based on their specificity and expression characteristics within the dataset.

Visual Summary

The dot plot effectively visualizes the expression of a broad panel of marker genes across 38 distinct celltype_subset populations. Key observations include:

• Clear Segregation of Cell Types: Most celltype_subset populations display highly specific and enriched expression of particular marker gene sets, forming distinct blocks along the diagonal, enclosed by red boxes. This indicates good separation and unique transcriptional identities for the annotated cell types.
• High Specificity: For many cell types, markers are expressed in a high percentage of cells within their respective group (large dot size) and at high mean expression levels (dark red color), with minimal or no expression in other cell types. This suggests robust and reliable markers for distinguishing these populations.
• Heterogeneity within Major Lineages: The plot successfully delineates various immune cell subsets (e.g., B cells, T cells, Macrophages, DCs, ILCs) and stromal cell subsets (e.g., Fibroblasts, Stellate cells, Smooth muscle cells), each with their unique marker signatures, confirming the fine-grained resolution of the celltype_subset annotation.

Examples of Highly Specific Markers:

• Acinar cells: Strongly express digestive enzyme genes like *CPB1, PRSS1, CPA1, PNLIP, SPINK1*.
• Ductal cells: Characterized by epithelial markers such as *KRT19, MUC1, CLDN4, TFF2*.
• Plasma cells: Show high expression of *JCHAIN, SDC1 (CD138), MZB1, XBP1*.
• Treg cells: Identified by classic markers *FOXP3, CTLA4, TNFRSF18*.
• Stellate cells: Marked by myofibroblast-associated genes like *ACTA2 (α-SMA), DCN, COL1A1*.
• Smooth muscle cells: Express contractile genes such as *ACTA2, MYH11, CNN1, TAGLN*.

Biological Interpretation

The observed marker expression patterns align strongly with established biological knowledge for these cell types in the human pancreas.

Pancreatic Epithelial Cells:

• Acinar cells exhibit robust expression of genes encoding digestive enzymes (e.g., *PRSS1* (Trypsin), *CPA1* (Carboxypeptidase A1), *PNLIP* (Pancreatic Lipase), *SPINK1* (Serine Protease Inhibitor Kazal Type 1)) [GeneCards: PRSS1, CPA1, PNLIP, SPINK1]. This confirms their role in exocrine function.
• Ductal cells, the presumed tumor origin cell type for PDAC, are characterized by cytokeratin *KRT19* [GeneCards: KRT19], mucin *MUC1* [GeneCards: MUC1], and trefoil factor family member *TFF2* [GeneCards: TFF2], consistent with their epithelial and secretory roles in duct lining.

Stromal Cells:

• Fibroblasts show high expression of extracellular matrix (ECM) components (*COL1A1, COL1A2, DCN, LUM*) and fibroblast activation protein (*FAP*) [GeneCards: FAP].
• Stellate cells, critical in pancreatic fibrosis, are identified by *ACTA2* (encoding α-smooth muscle actin, a marker of myofibroblast differentiation) [GeneCards: ACTA2], *DCN*, and collagen genes, reflecting their activated state and contribution to the tumor microenvironment.
• Smooth muscle cells are clearly defined by *ACTA2*, *MYH11* (Myosin Heavy Chain 11) [GeneCards: MYH11], and *TAGLN* (Transgelin) [GeneCards: TAGLN], indicative of their contractile phenotype.

Endothelial Cells:

• Endothelial tip cells express general endothelial markers (*CD34, VWF, PECAM1*) alongside specialized markers like *DLL4* [GeneCards: DLL4], involved in angiogenesis.
• Lymphatic Endothelial cells are distinguished by *PROX1* [GeneCards: PROX1] and *LYVE1* [GeneCards: LYVE1], key genes for lymphatic vessel identity.

Immune Cells:

• B cell subsets (Breg, Follicular, MZ, Memory) are collectively identified by pan-B cell markers (*CD79A, CD79B, MS4A1 (CD20)*) with distinct patterns for plasma cells (*JCHAIN, SDC1, XBP1, TNFRSF17* (BCMA)) [GeneCards: SDC1, TNFRSF17].
• Dendritic cell (DC) subsets (Classical, Inflammatory, Plasmacytoid) show unique markers such as *CD1C* for cDC [GeneCards: CD1C], *LILRA4* for pDC [GeneCards: LILRA4], and *CD86* for inflammatory DC.
• Macrophages are broadly characterized by *CD68* and exhibit distinct markers for M1-like (*TNF*) and M2-like polarization (*CD163, MSR1, SOCS3*) [GeneCards: CD163, MSR1], although M2 subtypes can have overlapping expression.
• Mast cells are unequivocally identified by *KIT* (CD117) [GeneCards: KIT] and *TPSAB1* (Tryptase Alpha/Beta 1) [GeneCards: TPSAB1].
• NK cells display cytotoxic granules (*NKG7, GZMB, PRF1*) [GeneCards: GZMB, PRF1] indicative of their immune surveillance role.

T cell subsets are well-resolved

• Cytotoxic T cells with *CD8A, CD8B, GZMB, PRF1*.
• Naive T cells with *SELL*.
• Treg cells with the master regulator *FOXP3* [GeneCards: FOXP3] and co-inhibitory receptor *CTLA4* [GeneCards: CTLA4].
• Helper T cell subsets (Th1, Th2, Th17, Th22) display lineage-specific transcription factors or effector molecules like *IFNG* (Th1), *GATA3* (Th2), and *RORC* (Th17/Th22).
• Innate Lymphoid Cells (ILCs) show markers like *EOMES* (ILC1) and *RORC* (ILC3), distinguishing them within the broader lymphoid compartment.

Annotation Notes

This analysis primarily serves as a robust quality control and validation step for the celltype_subset annotations. The high specificity and concordance of marker gene expression with known cellular identities strongly support the accuracy and reliability of the cell type assignments in this single-cell RNA-seq dataset. The clear separation of even closely related subsets (e.g., different T cell or macrophage subtypes) underscores the quality of the clustering and annotation process. This foundational validation is crucial for downstream analyses, ensuring that any biological conclusions drawn are based on accurately identified cell populations.

4. Ductal Cell Copy Number Variation (CNV) Analysis in Pancreatic Cancer

[Analysis Visualization Results]...

Analysis Overview

This analysis investigates copy number variations (CNVs) specifically within Ductal cells from human pancreas tissue, comparing adjacent normal (Adj_normal) and Pancreatic Ductal Adenocarcinoma (PDAC) conditions. Ductal cells are identified as the tumor origin cell type in this dataset. The provided visualizations include a CNV heatmap showing log2(Copy Number Ratio, CNR) values across genomic spots for individual samples, grouped by cell group (sample and ploidy status), and a summary heatmap with a corresponding bar plot highlighting frequently amplified cytogenetic bands and their average amplification levels across PDAC samples. These results help characterize the genomic landscape of Ductal cells in PDAC and identify recurrent CNV events.

Visual Summary

CNV Heatmap (log2(CNR))

The heatmap displays log2(CNR) values, where red indicates amplification (log2(CNR) > 0) and blue indicates deletion (log2(CNR) < 0).

• Distinct Patterns by Condition: Cells from Adj_normal samples (AdjN_1, AdjN_2, AdjN_3) exhibit a largely uniform, flat blue pattern across all chromosomes, indicating a near-diploid state with minimal CNVs, as expected for healthy tissue. In stark contrast, PDAC samples show widespread and diverse CNV patterns.
• Heterogeneity within PDAC: There is significant heterogeneity in CNV profiles among PDAC samples. While some PDAC samples (e.g., PDAC_1, PDAC_13, PDAC_15, PDAC_16) display extensive amplifications and deletions across multiple chromosomes, others labeled as "Diploid PDAC" (e.g., Diploid PDAC_10, Diploid PDAC_11A/B, Diploid PDAC_2-6, Diploid PDAC_8-9) generally show fewer, often more focal, CNVs. This distinction suggests that while overall ploidy might be near diploid, regional CNVs are still prevalent in these "Diploid PDAC" samples.
• Recurrent Amplifications: Several chromosomal regions show recurrent amplifications across multiple PDAC samples, notably:
• Chromosome 1q: Broad amplification observed in many PDAC samples, including regions like 1q23.2.
• Chromosome 7p/q: A distinct amplification peak around chromosome 7 (e.g., 7p13:7q21.11), which is known to harbor the *EGFR* gene. This amplification is prominent in samples like PDAC_1, 13, 15, 16, 2, 3, 5, 6, 7.
• Chromosome 8q: Amplification at 8q (e.g., 8q22.1:8q24.3) is also frequently observed.
• Chromosome 17q: Amplification around 17q12:17q21 is seen in several samples.
• Chromosome 19q: A recurrent amplification at 19q13:19q13.31.
• Chromosome 20q: Amplification at 20q13.
• Recurrent Deletions: Fewer extensive deletions are visually prominent compared to amplifications, but some focal deletions are visible, for example, on chromosomes 6q and 9p in some PDAC samples.

Significantly Amplified Regions Summary

The summary plot quantifies the most frequently amplified cytogenetic bands and their average log2(CNR) values across the PDAC samples.

Top Amplified Regions (Frequency)

• The most frequently amplified region is 1q21.3:1q23.2, observed in 82% of PDAC samples.
• Following closely is 1q23.2:2p25.2, also at 82% frequency.
• 7p13:7q21.11 (EGFR) is amplified in 45% of PDAC samples.
• 19q13.2:19q13.31 is amplified in 45% of PDAC samples.
• Other notable frequent amplifications include 3q21.3:3q25.1, 3q27.1:3q29, and 8q22.1:8q24.3 (containing *EIF3E, GSDMD*).

Amplification Magnitude (Average log2(CNR))

• Sample PDAC_1 shows high average log2(CNR) values across several amplified regions, particularly 1q21.3:1q23.2 (2.4), 1q23.2:2p25.2 (0.7), 7p13:7q21.11 (1.2), and 19q13.2:19q13.31 (1.3).
• PDAC_8 also exhibits strong amplifications in 1q21.3:1q23.2 (1.0), 1q23.2:2p25.2 (1.2), and 8q22.1:8q24.3 (1.2).
• The values in the summary heatmap indicate the average log2(CNR) within that specific cytogenetic band for each sample. Values of '0.0' indicate no significant amplification in that region for that sample.

Biological Interpretation

The CNV analysis of Ductal cells clearly distinguishes between non-malignant and malignant pancreatic tissues. The near absence of CNVs in Adj_normal Ductal cells confirms their healthy genomic state, serving as a robust reference. In contrast, PDAC Ductal cells exhibit extensive and highly heterogeneous CNV profiles, which is a hallmark of cancer genomes and particularly aggressive tumors like PDAC.

• Tumor-specific Genomic Alterations: The widespread CNVs observed exclusively in PDAC samples strongly indicate their oncogenic nature. As Ductal cells are the "Tumor origin celltype," these CNVs likely represent early and sustained genomic alterations driving tumor initiation and progression.
• Role of Ploidy: The distinction between "Diploid PDAC" and other "PDAC" samples suggests that the ploidy_dec annotation reflects the overall chromosomal complement rather than the complete absence of segmental CNVs. Even "Diploid PDAC" samples demonstrate focal amplifications, indicating that genomic instability leading to CNVs can occur independently of gross whole-genome duplication events. This highlights the complexity of tumor ploidy and CNV landscapes.
• Key Oncogenic Drivers: The recurrent amplification of specific cytogenetic bands points to regions containing genes critical for PDAC pathogenesis.
• 7p13:7q21.11 (EGFR): Amplification of the *EGFR* locus is highly significant. The Epidermal Growth Factor Receptor (EGFR) is a well-known oncogene that promotes cell proliferation, survival, and metastasis in various cancers, including a subset of PDACs. Its amplification suggests an activation of the EGFR signaling pathway, which can drive tumor growth UniProt: P00533.
• 1q and 8q Amplifications: Recurrent amplifications on chromosomes 1q and 8q are frequently observed in various cancers, including PDAC. These regions often harbor multiple oncogenes. For example, 8q24.3 is known to contain *MYC*, a powerful oncogene, though it's not explicitly named here, *EIF3E* and *GSDMD* are mentioned. *EIF3E* (Eukaryotic Translation Initiation Factor 3 Subunit E) is implicated in regulating protein synthesis and has been found to be overexpressed or amplified in various cancers, contributing to oncogenesis GeneCards: EIF3E. *GSDMD* (Gasdermin D) plays a role in pyroptosis, a form of programmed cell death, and its dysregulation via CNV could impact immune evasion or tumor progression GeneCards: GSDMD.
• Intra-tumor Heterogeneity: The variability in CNV patterns across different PDAC samples underscores the substantial intra-tumor heterogeneity characteristic of pancreatic cancer. This genomic diversity can have profound implications for treatment response and resistance.

Clinical or Translational Implications

The identification of recurrent and specific CNV patterns in Ductal cells of PDAC samples has several clinical and translational implications:

• Biomarker Potential: These specific CNVs, particularly amplifications of regions like 1q, 7p (EGFR), 8q, and 19q, could serve as diagnostic or prognostic biomarkers for PDAC. Detecting these genomic alterations in patient samples could aid in early diagnosis or predict disease aggressiveness.
• Therapeutic Targets: The amplification of *EGFR* is particularly actionable. Patients with *EGFR* amplification might be candidates for targeted therapies with EGFR inhibitors (e.g., erlotinib, gefitinib), which are already utilized in other cancer types. This highlights a potential avenue for personalized medicine in a subset of PDAC patients, although resistance mechanisms are common and would need to be considered PubMed: EGFR inhibitors pancreatic cancer. Further investigation into the functional consequences of *EIF3E* or *GSDMD* amplifications could also reveal novel therapeutic targets.
• Understanding Tumor Evolution: The observed genomic heterogeneity among PDAC samples, even within the "Diploid PDAC" group, emphasizes the dynamic nature of tumor evolution. Monitoring CNV changes could provide insights into disease progression, metastasis, and the development of treatment resistance, informing adaptive treatment strategies.
• Precision Medicine: The ability to profile CNVs at a single-cell level within the tumor-origin cell type (Ductal cells) allows for a more granular understanding of individual tumor characteristics, paving the way for more precise and effective patient stratification and treatment selection.

5. UMAP Visualization of CNV Patterns Across Pancreatic Cell Types and Conditions

[Analysis Visualization Results]...

Analysis Overview

This analysis presents UMAP (Uniform Manifold Approximation and Projection) visualizations of single-cell RNA-seq data, where the dimensionality reduction is performed based on estimated Copy Number Variation (CNV) patterns (from obsm['X_cnv']). The resulting UMAP plots are colored by different categorical annotations including major cell type (celltype_major), minor cell type (celltype_minor), ploidy status (ploidy_dec), disease condition (condition), and individual sample (sample). This visualization helps to understand the underlying CNV heterogeneity, its association with cell identity, ploidy status, disease state, and potential sample-specific effects.

Visual Summary

The UMAPs reveal distinct clustering patterns driven by CNV profiles, with clear separation of cells based on their genomic integrity.

• Ploidy Status (ploidy_dec): The most striking feature is the distinct segregation of cells based on their inferred ploidy. A large, well-defined cluster of 'Aneuploid' cells (maroon) is clearly separated from the majority 'Diploid' cells (yellow). A very small number of 'Unclear' cells are also present but do not form a distinct cluster. This strong separation indicates that CNV patterns are a major driver of the embedding structure.
• Major Cell Type (celltype_major): The 'Aneuploid' cluster predominantly consists of 'Ductal cells' (orange/red-orange). 'Acinar cells' (maroon) also show some presence within or adjacent to the aneuploid regions, though primarily they reside in the diploid cluster. All other major cell types (e.g., Stromal cell, Endothelial cell, T cell, Myeloid cell, B cell, Mast cell) are almost exclusively found within the 'Diploid' clusters, as expected for non-malignant cells.
• Minor Cell Type (celltype_minor): Refining the celltype_major view, the 'Aneuploid' cluster is strongly enriched for 'Ductal cells' (orange). Other stromal cells like 'Stellate cell' and 'Fibroblast' (light blue and cream, respectively) primarily co-cluster with other diploid cells. This further confirms the strong association of aneuploidy with the putative malignant epithelial cell population.
• Condition (condition): The 'Aneuploid' clusters are almost exclusively derived from the 'PDAC' (Pancreatic Ductal Adenocarcinoma, purple) condition. Conversely, 'Adj_normal' (Adjacent Normal, maroon) cells are predominantly found in the 'Diploid' clusters. This strongly correlates aneuploidy with the disease state. It's important to note that many 'PDAC' cells are also present in the 'Diploid' clusters, which would represent the tumor microenvironment components (e.g., stromal, immune cells) from PDAC samples.
• Sample (sample): The 'Diploid' clusters show a relatively mixed distribution of cells from various 'Adj_normal' and 'PDAC' samples, suggesting shared CNV landscapes among non-malignant cells and a lack of strong batch effects dominating the diploid cell embedding. In contrast, within the 'Aneuploid' region, while several 'PDAC' samples contribute, there are also visually discernible sub-clusters or regions dominated by specific 'PDAC' samples (e.g., PDAC_1, PDAC_2, PDAC_3, etc.). This indicates significant inter-patient heterogeneity in the CNV profiles of malignant cells.

Biological Interpretation

The UMAP visualizations of CNV patterns provide compelling biological insights into Pancreatic Ductal Adenocarcinoma (PDAC) pathogenesis and the composition of the tumor microenvironment.

1. Clear Demarcation of Malignant Cells: The most significant finding is the robust separation of aneuploid cells, which are overwhelmingly identified as Ductal cells from PDAC samples. This strongly supports the notion that these aneuploid Ductal cells represent the malignant epithelial compartment of PDAC, characterized by significant genomic instability, a hallmark of cancer 1.
2. Origin of Aneuploidy: Given that Ductal cells are identified as the 'Tumor origin celltype' in the data context, the observed high prevalence of aneuploidy specifically in Ductal cells from PDAC samples aligns perfectly with the known biology of PDAC, which typically originates from the ductal epithelium 2.
3. Heterogeneity within PDAC: The presence of distinct sample-specific sub-clusters within the aneuploid Ductal cell population suggests significant inter-tumor heterogeneity in CNV landscapes among different PDAC patients. This is a critical aspect of cancer biology, impacting patient response to therapy and disease progression.
4. Tumor Microenvironment Integrity: The stromal and immune cells (e.g., Stellate, Fibroblast, T cells, Macrophages, Endothelial cells) from both 'Adj_normal' and 'PDAC' conditions primarily cluster within the diploid regions. This indicates that their global CNV profiles remain relatively stable and diploid, distinguishing them from the malignant aneuploid cells. While these cells play crucial roles in supporting tumor growth and immune response, their genomic integrity, as assessed by CNV, appears largely preserved.
5. Acinar Cell Observations: The minor presence of Acinar cells within or near the aneuploid regions warrants further investigation. While PDAC predominantly arises from ductal cells, acinar-to-ductal metaplasia (ADM) is a known precursor lesion, and some forms of pancreatic cancer can have acinar origins or features. This observation could potentially reflect these transitional states or less common tumor origins.

Clinical or Translational Implications

These CNV-based UMAPs serve as a powerful validation of cell type assignments and provide a foundational understanding of the genomic landscape of PDAC.

• Diagnostic Potential: The clear distinction between aneuploid (malignant) and diploid (non-malignant) cells based on CNV could have future implications for early diagnostic markers or confirming tumor cell presence in biopsy samples, especially when combined with cell type-specific markers.
• Therapeutic Stratification: The observed inter-patient heterogeneity in CNV patterns among PDAC cells highlights the need for personalized medicine approaches. Understanding these diverse genomic alterations could help in stratifying patients for targeted therapies.
• Research Model Validation: The ability to clearly separate tumor cells from the tumor microenvironment based on CNV profiles validates the quality of the single-cell data and the CNV inference method, providing a robust framework for subsequent downstream analyses such as differential gene expression or cell-cell interaction studies, ensuring these analyses are performed on biologically distinct populations.

---

References:

1. Genomic Instability in Cancer:

PubMed search: Genomic instability cancer

2. Pancreatic Cancer Origin:

PubMed search: Pancreatic cancer ductal origin

6. Cell Type Population Analysis of Minor Cell Types in Pancreatic Samples

[Analysis Visualization Results]...

Analysis Overview

This analysis provides a population bar plot illustrating the proportional distribution of minor cell types across individual samples from both 'Adj_normal' (adjacent normal pancreas) and 'PDAC' (Pancreatic Ductal Adenocarcinoma) conditions. The plot allows for a visual comparison of cell type composition shifts associated with PDAC development and progression, highlighting differences in the cellular microenvironment between healthy and cancerous pancreatic tissue.

Visual Summary

The bar plot displays the relative abundance of 19 distinct minor cell types within each sample. Samples are grouped by condition: 'Adj_normal' (3 samples: AdjN_3, AdjN_1, AdjN_2) and 'PDAC' (14 samples).

• Acinar Cell Depletion in PDAC: 'Acinar cells' (dark red) constitute a substantial proportion of the Adj_normal samples, as expected for healthy pancreatic tissue. In stark contrast, their proportion is drastically reduced, or nearly absent, across almost all PDAC samples, indicating replacement of normal pancreatic parenchyma by tumor and stromal components.
• Ductal Cell Expansion in PDAC: 'Ductal cells' (orange), which are the tumor origin cell type, show a marked increase in proportion in many PDAC samples (e.g., PDAC_16, PDAC_6, PDAC_3, PDAC_8, PDAC_2, PDAC_7, PDAC_13), often becoming a dominant cell type. This likely includes both malignant ductal cells and possibly reactive ductal cells.
• Stromal Cell Accumulation in PDAC: There is a noticeable enrichment of stromal cell populations in PDAC samples.
• 'Fibroblasts' (light orange) are present in both conditions but appear more prominent in many PDAC samples.
• 'Stellate cells' (light green), a key component of the desmoplastic reaction in PDAC, are consistently present in PDAC samples, often forming a significant proportion (e.g., PDAC_16, PDAC_6, PDAC_3, PDAC_8), while being less prominent or absent in Adj_normal.
• 'Endothelial cells' (peach) also appear to contribute more substantially to the cellular landscape in PDAC, likely supporting tumor vascularization.
• Immune Cell Infiltration: The immune cell compartment shows significant alterations in PDAC.
• 'Macrophages' (pale yellow) appear enriched in a large proportion of PDAC samples compared to Adj_normal, consistent with their role in the tumor microenvironment.
• 'T cells CD4+' (teal) and 'T cells CD8+' (dark teal) are present in both conditions but show variable proportions across PDAC samples, reflecting the diverse immune responses and immune evasion mechanisms in different tumors.
• 'B cells' (red) and 'Plasma cells' (light green) are also observed, with their proportions varying across PDAC samples.
• 'Dendritic cells', 'ILC', 'Mast cells', and 'NK cells' are also present, generally in smaller proportions, with some variability.
• Inter-sample Heterogeneity in PDAC: There is considerable heterogeneity in cell type composition among individual PDAC samples, reflecting the complex and diverse nature of pancreatic cancer. For example, some PDAC samples (e.g., PDAC_16, PDAC_3) are highly enriched in Ductal cells and Stellate cells, while others (e.g., PDAC_1A, PDAC_5, PDAC_10, PDAC_15) show a higher proportion of immune cells like Macrophages.
• Unassigned Cells: The proportion of 'unassigned' cells (dark blue) is relatively low across all samples, indicating good cell type annotation quality.

Biological Interpretation

The observed shifts in cell type populations are highly characteristic of the pancreatic ductal adenocarcinoma tumor microenvironment (TME).

• Loss of Normal Pancreatic Architecture: The drastic reduction of Acinar cells and the concomitant increase in Ductal cells (including malignant cells) in PDAC samples reflects the neoplastic transformation and replacement of healthy pancreatic tissue by the tumor.
• Desmoplastic Stroma Formation: The significant expansion of Fibroblasts and Stellate cells in PDAC samples underscores the prominent desmoplastic reaction, a hallmark of PDAC. Pancreatic stellate cells (PSCs) are key orchestrators of this fibrotic reaction, secreting extracellular matrix components that contribute to tumor hardness and create a physical barrier impeding drug delivery and immune cell infiltration [PubMed Search: pancreatic stellate cells PDAC stroma].
• Immune Dysregulation in the TME: The altered immune cell landscape, particularly the enrichment of Macrophages, points towards immune modulation within the PDAC TME. Tumor-associated macrophages (TAMs), especially M2-like macrophages, are known to promote tumor growth, angiogenesis, and immune suppression in PDAC [GeneCards: CD68 (macrophage marker)]. The varying proportions of T cells (CD4+ and CD8+) suggest diverse immune responses, which can range from active immune surveillance to profound immunosuppression mediated by the TME.
• Tumor-associated Angiogenesis: The increased presence of Endothelial cells likely reflects tumor-driven angiogenesis, which is essential for tumor growth and metastasis.

Clinical or Translational Implications

The distinct cellular composition of PDAC samples compared to adjacent normal tissue offers several clinical and translational insights:

• Diagnostic Potential: The significant shift from Acinar cells to Ductal cells (malignant), combined with an increase in stromal and specific immune populations, could serve as a diagnostic signature for PDAC, potentially informing biopsy interpretation or liquid biopsy development [PubMed Search: PDAC cell composition biomarker].
• Therapeutic Targeting of the TME: The prominence of Stellate cells and Fibroblasts highlights the stroma as a critical therapeutic target. Strategies aimed at depleting or reprogramming PSCs could improve drug penetration and enhance the efficacy of chemotherapy or immunotherapy [PubMed Search: PDAC stroma targeting therapy].
• Immunotherapy Strategies: The altered immune landscape, especially the infiltration of Macrophages, suggests that therapies targeting TAMs or strategies to reprogram them (e.g., from M2 to M1 phenotype) could be beneficial. Understanding the balance and state of T cell populations (CD4+ vs. CD8+, effector vs. regulatory) is crucial for designing effective immunotherapies for PDAC, which has historically been resistant to such treatments [UniProt: P01897 (CD4), P01742 (CD8A)].
• Prognostic Value: The specific cellular proportions and their heterogeneity among PDAC samples may hold prognostic value, where certain cell type enrichments (e.g., high TAMs or specific stromal signatures) could correlate with disease aggressiveness or treatment response.

7. T cell Subtype Population Analysis in Pancreatic Ductal Adenocarcinoma (PDAC)

[Analysis Visualization Results]...

Analysis Overview

This analysis visualizes the proportional distribution of T cell subsets, including Innate Lymphoid Cells (ILCs) and NK cells, within the T cell major cell type across individual samples from both "Adj_normal" (adjacent normal pancreas tissue) and "PDAC" (Pancreatic Ductal Adenocarcinoma) conditions. The celltype_subset annotation level provides a granular view of these immune populations. This population bar plot helps to understand the shifts in immune cell composition within the tumor microenvironment compared to healthy tissue.

Visual Summary

The visualization presents two groups of stacked bar plots: one for "Adj_normal" samples and another for "PDAC" samples. Each bar represents a distinct sample, with segment heights indicating the relative proportion of each T cell subset (and related ILCs/NK cells) within that sample.

• Adj_normal Samples: The three "Adj_normal" samples (AdjN_3, AdjN_2, AdjN_1) show a relatively consistent composition. These samples are predominantly composed of T cell (Naive), T cell (Cytotoxic), and T cell (Tfh) populations, with very low proportions of ILCs, NK cells, and other T cell subsets.
• PDAC Samples: The "PDAC" samples exhibit a more diverse and heterogeneous immune landscape compared to the adjacent normal tissue.
• Increased Innate Lymphoid Cells (ILCs) and NK cells: There is a noticeable increase in the proportions of ILC1, ILC2, ILC3 (NCR+), ILC3 (NCR-), and NK cells across many PDAC samples. ILC1 (dark red) is particularly prominent in several PDAC samples (e.g., PDAC_11A, PDAC_16, PDAC_I3).
• Presence of Regulatory T cells (Tregs): T cell (Treg) populations (darker blue) are more consistently observed in PDAC samples, whereas they are almost absent in Adj_normal samples.
• Shift in T cell subtypes: While T cell (Cytotoxic) and T cell (Naive) remain significant components, their proportional representation appears reduced in some PDAC samples compared to Adj_normal, relative to the increased presence of other cell types. Other T helper subsets like T cell (Th17), T cell (Th1), T cell (Th2), T cell (Th9), and T cell (Th22) are present in varying, generally smaller, proportions across PDAC samples.
• Inter-sample heterogeneity: Significant variability in the proportions of these immune cell subsets is observed among different PDAC patients, highlighting the diverse immune responses within the PDAC tumor microenvironment.

Biological Interpretation

The observed shifts in immune cell populations between adjacent normal pancreas and PDAC tumors provide key insights into the immunobiology of pancreatic cancer.

1. Innate Lymphoid Cell (ILC) Infiltration: The significant increase in various ILC subsets (ILC1, ILC2, ILC3) in PDAC samples suggests a robust involvement of the innate immune system in the tumor microenvironment.

• ILC1s are typically associated with Th1-type immune responses and IFN-γ production, often linked to anti-tumor immunity. Their increased presence might reflect an attempt by the host immune system to combat the tumor [PubMed search: ILC1 cancer].
• Conversely, ILC2s are known to promote Th2-type inflammation, tissue remodeling, and fibrosis, which can contribute to an immunosuppressive and pro-tumorigenic microenvironment in PDAC [PubMed search: ILC2 PDAC].
• ILC3s, producers of IL-17 and IL-22, have complex roles, potentially promoting inflammation that can either restrict or support tumor growth depending on context [PubMed search: ILC3 tumor immunology]. The co-existence of different ILC subsets suggests a complex interplay of immune responses.

2. NK Cell Dynamics: The increased presence of NK cells in PDAC samples is notable. NK cells are critical for direct cytolysis of tumor cells and secretion of anti-tumor cytokines [GeneCards: NCR1, UniProt: O95914]. Their enhanced presence could indicate an active anti-tumor immune surveillance, though their function can be impaired in the tumor microenvironment.
3. Regulatory T cell (Treg) Expansion: The clear emergence of T cell (Treg) populations in PDAC, virtually absent in normal tissue, is a well-documented phenomenon in many cancers, including PDAC [PubMed search: Treg PDAC]. Tregs suppress anti-tumor immune responses, contributing to immune evasion and tumor progression [GeneCards: FOXP3, UniProt: P43058].
4. Adaptive T cell Remodeling: While T cell (Cytotoxic) cells are vital for anti-tumor immunity, their relative proportional reduction (compared to other expanded subsets) in some PDAC samples might imply an overwhelmed or suppressed cytotoxic response. The presence of diverse T helper subsets (Th1, Th2, Th17, Th22, Th9) further underscores the complex adaptive immune response within the tumor, which can have both pro- and anti-tumorigenic implications depending on the dominant polarization. For instance, Th17 cells have been shown to promote tumor progression in PDAC by fostering inflammation and angiogenesis [PubMed search: Th17 pancreatic cancer].

Clinical or Translational Implications

The distinct shifts in T cell and ILC populations between normal and PDAC tissues offer several potential clinical and translational implications:

• Prognostic Biomarkers: The specific composition and ratios of ILCs, NK cells, and various T cell subsets (e.g., ILC2/ILC1 ratio, Treg frequency) could serve as prognostic biomarkers for disease progression and patient outcomes in PDAC.

Immunotherapeutic Targets:

• The increased presence of immunosuppressive cells like Tregs suggests that therapies aimed at depleting or inhibiting Treg function could enhance anti-tumor immunity in PDAC [PubMed search: Treg depletion cancer therapy].
• Modulating the activity of specific ILC subsets (e.g., inhibiting pro-tumorigenic ILC2s or enhancing anti-tumorigenic ILC1s) could represent novel immunotherapeutic strategies.
• Strategies to enhance NK cell function or infiltration in PDAC could also be explored to boost anti-tumor responses.
• Patient Stratification: The observed inter-patient heterogeneity in immune cell composition within PDAC samples highlights the need for personalized immunotherapy approaches. Patients could potentially be stratified based on their tumor immune microenvironment profiles to guide treatment decisions.

8. Differential T cell Subset Proportions in Pancreatic Ductal Adenocarcinoma (PDAC)

[Analysis Visualization Results]...

Analysis Overview

This analysis investigates the proportions of various T cell subsets within single-cell RNA-seq data from human pancreas tissue, comparing samples from Pancreatic Ductal Adenocarcinoma (PDAC) patients to adjacent normal (Adj_normal) tissue. The objective is to identify T cell subsets that show statistically significant differences in their relative abundances between these two conditions, providing insights into the immune microenvironment of PDAC.

Visual Summary

The box plots display the proportional representation of eight T cell subsets (T_Cyto, T_Naive, Treg, Th9, Th17, Tfh, Th2, Th22) across PDAC and Adj_normal conditions. For each T cell subset, a p-value indicates the statistical significance of the difference between the two conditions.

• T_Cytotoxic (T_Cyto): Significantly *lower* in PDAC (median approx. 30%) compared to Adj_normal (median approx. 80%) (p=6.75e-11). This represents a stark decrease in cytotoxic T cells within the tumor environment.
• T_Naive: Significantly *higher* in PDAC (median approx. 28%) compared to Adj_normal (median approx. 2.5%) (p=5.39e-07). This suggests an enrichment of undifferentiated T cells in the tumor.
• Treg (Regulatory T cells): Significantly *higher* in PDAC (median approx. 5%) compared to Adj_normal (median approx. 1%) (p=0.000659). This indicates an increased presence of immunosuppressive T cells in PDAC.
• Th9: Significantly *higher* in PDAC (median approx. 0.4%) compared to Adj_normal (median approx. 0%) (p=0.00114).
• Th17: Significantly *higher* in PDAC (median approx. 2%) compared to Adj_normal (median approx. 0.2%) (p=0.00235).
• Tfh (Follicular Helper T cells): Significantly *higher* in PDAC (median approx. 7.5%) compared to Adj_normal (median approx. 2.5%) (p=0.00266).
• Th2: Significantly *higher* in PDAC (median approx. 1%) compared to Adj_normal (median approx. 0.1%) (p=0.00406).
• Th22: Significantly *higher* in PDAC (median approx. 2.5%) compared to Adj_normal (median approx. 1%) (p=0.00596).

In summary, all T cell subsets displayed, except for Cytotoxic T cells, are significantly more abundant in PDAC compared to adjacent normal tissue. Cytotoxic T cells are significantly *reduced* in PDAC.

Biological Interpretation

The observed shifts in T cell subset proportions between PDAC and adjacent normal tissue reveal a distinct immunological landscape within the tumor microenvironment (TME) of pancreatic cancer.

1. Reduced Anti-Tumor Immunity: The most striking finding is the significantly *lower proportion of Cytotoxic T cells (T_Cyto)* in PDAC. Cytotoxic T cells are critical effector cells responsible for directly recognizing and killing tumor cells. Their scarcity or functional impairment in the PDAC TME is a hallmark of immune evasion, directly contributing to tumor progression and resistance to immunotherapy [1].
2. Increased Immunosuppression: The *significant increase in Regulatory T cells (Tregs)* in PDAC is a major indicator of an immunosuppressive TME. Tregs actively suppress anti-tumor immune responses by inhibiting the proliferation and function of effector T cells and other immune cells, thereby promoting tumor growth and metastasis [2].
3. Accumulation of Naive T cells: The *higher proportion of Naive T cells* in PDAC suggests that while T cells are present, many may not be effectively activated or differentiated into tumor-specific effector cells. This could be due to a lack of proper antigen presentation, insufficient co-stimulation, or the overwhelming immunosuppressive signals within the PDAC TME, preventing productive anti-tumor immunity.
4. Complex Roles of Helper T cell Subsets:

• Th17 cells are increased in PDAC. Th17 cells have context-dependent roles in cancer; they can be pro- or anti-tumorigenic. In PDAC, they are often associated with chronic inflammation, tumor progression, and metastasis, potentially by promoting angiogenesis and recruitment of other immune cells [3].
• Th2 cells are also elevated. Th2 responses are generally associated with humoral immunity and allergic reactions, but in cancer, they can contribute to an immunosuppressive microenvironment by promoting M2 macrophage polarization and inhibiting Th1 responses [4].
• Th9, Th22, and Tfh cells are likewise enriched in PDAC. The roles of these subsets in cancer, particularly PDAC, are still being actively investigated.
• Th9 cells can exert both anti-tumor and pro-tumor effects depending on the cancer type and microenvironment. Their presence in PDAC might contribute to immune modulation.
• Th22 cells are involved in epithelial immunity and inflammation. In some cancers, they are linked to tumor progression, angiogenesis, and resistance to therapy.
• Tfh cells are crucial for B cell maturation and antibody production within germinal centers. Their increased presence might reflect the formation of tertiary lymphoid structures or an atypical humoral response within the PDAC TME, the full implications of which are not always clear in promoting anti-tumor immunity.

Overall, the T cell landscape in PDAC appears to be skewed towards an immunosuppressive and potentially pro-tumorigenic profile, characterized by a deficiency of cytotoxic T cells and an abundance of regulatory and various helper T cell subsets that may either contribute to immune suppression or dysfunctional immune responses.

Clinical or Translational Implications

The distinct T cell subset profiles in PDAC highlight several potential clinical and translational implications:

1. Biomarker Potential: The proportions of these T cell subsets, particularly the low T_Cyto/Treg ratio, could serve as prognostic biomarkers for PDAC progression or indicators of response to immunotherapy.
2. Therapeutic Targets: The observed imbalance suggests several therapeutic strategies:

• Enhancing Cytotoxic T cell Function: Strategies to recruit, activate, and maintain the function of cytotoxic T cells within the PDAC TME are crucial.
• Targeting Immunosuppression: Depleting or inhibiting the function of Tregs could unleash anti-tumor immunity.
• Modulating Helper T cells: Understanding the precise pro-tumor functions of Th2, Th9, Th17, Th22, and Tfh cells in PDAC could lead to therapies aimed at reprogramming their activity or reducing their numbers.

3. Rationale for Immunotherapy Resistance: The findings align with the known resistance of PDAC to current immunotherapies. The TME appears to be highly immune-privileged, actively suppressing effective anti-tumor responses through the observed shifts in T cell populations. Overcoming this immunosuppression will be key to improving treatment outcomes for PDAC patients.

References:

1. Cytotoxic T cells in cancer: PubMed search for "cytotoxic T cells cancer immunity" https://pubmed.ncbi.nlm.nih.gov/?term=cytotoxic+T+cells+cancer+immunity
2. Regulatory T cells in cancer: PubMed search for "regulatory T cells tumor immunosuppression" https://pubmed.ncbi.nlm.nih.gov/?term=regulatory+T+cells+tumor+immunosuppression
3. Th17 cells in PDAC: PubMed search for "Th17 pancreatic cancer prognosis" https://pubmed.ncbi.nlm.nih.gov/?term=Th17+pancreatic+cancer+prognosis
4. Th2 cells in cancer: PubMed search for "Th2 cells cancer immunosuppression" https://pubmed.ncbi.nlm.nih.gov/?term=Th2+cells+cancer+immunosuppression

9. Ductal Cell Ploidy Analysis in Pancreatic Adenocarcinoma (PDAC)

[Analysis Visualization Results]...

Analysis Overview

This analysis investigates the ploidy status (Aneuploid vs. Diploid) of Ductal cells, which are identified as the tumor origin cell type, across different samples and conditions (Adj_normal vs. PDAC). The results are presented as stacked bar plots showing the proportional distribution of aneuploid, diploid, and unclear cells within Ductal cell populations for each sample.

Visual Summary

The visualization displays the ploidy profiles of Ductal cells across three adjacent normal (Adj_normal) samples and fourteen pancreatic ductal adenocarcinoma (PDAC) samples.

• Adjacent Normal Samples (Adj_normal): In all three Adj_normal samples (AdjN_1, AdjN_2, AdjN_3), Ductal cells are predominantly diploid (represented in orange), accounting for approximately 95-99% of the population. A very small fraction (<5%) appears aneuploid (maroon), and the "Unclear" category (light green) is almost negligible. This pattern suggests that normal Ductal cells typically maintain a diploid state.
• PDAC Samples: In stark contrast, Ductal cells from PDAC samples exhibit a high degree of aneuploidy.
• A significant proportion of PDAC samples (e.g., PDAC_13, PDAC_16, PDAC_2, PDAC_3, PDAC_15, PDAC_1, PDAC_6) show Ductal cell populations that are largely aneuploid, often exceeding 80-90%.
• There is considerable heterogeneity among PDAC samples. Some samples (e.g., PDAC_7, PDAC_8, PDAC_5) display a more mixed ploidy profile, with a substantial portion of Ductal cells remaining diploid (e.g., PDAC_8 shows roughly 50% diploid, 50% aneuploid).
• A few PDAC samples (e.g., PDAC_9, PDAC_12, PDAC_11B, PDAC_10, PDAC_4, PDAC_11A) show Ductal cells that are predominantly diploid, similar to the Adj_normal samples, or with only a minor aneuploid fraction. The "Unclear" category remains minimal across most PDAC samples.

Biological Interpretation

The observed shift in ploidy status in Ductal cells from Adj_normal to PDAC conditions provides critical biological insights into pancreatic cancer development.

• Aneuploidy as a Hallmark of Cancer: The dramatically increased proportion of aneuploid Ductal cells in many PDAC samples is a strong indicator of malignant transformation. Aneuploidy, defined as an abnormal number of chromosomes, is a well-established hallmark of cancer, contributing to genomic instability and driving tumor evolution. This finding aligns with Ductal cells being the tumor origin cell type in PDAC, as their malignant transformation would involve such genomic alterations. PubMed search: Aneuploidy cancer hallmark
• Heterogeneity in PDAC: The significant sample-to-sample variability in ploidy within the PDAC group highlights the known biological heterogeneity of pancreatic cancer. This could reflect:
• Tumor purity: Samples with a higher proportion of diploid Ductal cells might have a larger admixture of non-malignant, stromal, or immune cells, or even residual normal Ductal cells within the tumor microenvironment. However, since the analysis specifically targets "Ductal cells," this suggests heterogeneity *within* the Ductal cell compartment itself.
• Disease stage or progression: Tumors might acquire aneuploidy at different rates or stages of progression. Some tumors might be early-stage or less aggressive, retaining a more diploid state, while others are highly advanced and overtly aneuploid.
• Subclonal evolution: Even within a single tumor, different subclones can exist with varying degrees of aneuploidy, and the specific biopsy or single-cell sampling might capture these different populations.
• Technical considerations: While less likely to explain such large differences, technical factors in CNV estimation or cell calling could contribute to minor variations.
• Ductal Cell Malignancy: The observation of high aneuploidy specifically in Ductal cells from PDAC samples underscores their direct involvement in tumor initiation and progression. This genetic instability in the presumed cell of origin is a foundational characteristic of carcinogenesis.

Clinical or Translational Implications

• Biomarker Potential: The presence of a high proportion of aneuploid Ductal cells could serve as a valuable diagnostic or prognostic biomarker for PDAC. Detecting increased aneuploidy in pancreatic ductal cells, perhaps from biopsy or liquid biopsy samples, could indicate malignancy or a more aggressive disease state.
• Therapeutic Targeting: The underlying mechanisms driving aneuploidy (e.g., defects in chromosome segregation, cell cycle checkpoints) are potential therapeutic targets. Understanding the specific genomic alterations leading to aneuploidy in different PDAC subtypes could pave the way for personalized therapeutic strategies.
• Disease Monitoring: Monitoring the ploidy status of Ductal cells could potentially be used to track disease progression or response to treatment. A decrease in aneuploid cell populations after therapy might indicate a positive response.

10. Pancreatic Ductal Adenocarcinoma (PDAC) Cell-Cell Interaction Landscape: Shifts Towards Immunosuppression and Pro-Tumorigenic Signaling

[Analysis Visualization Results]...

Analysis Overview

This analysis investigates cell-cell interaction (CCI) patterns in pancreatic tissue, comparing adjacent normal pancreas (Adj_normal) with Pancreatic Ductal Adenocarcinoma (PDAC). Using single-cell RNA sequencing data and CellPhoneDB, we focused on interactions involving key cell types: Ductal cells (the tumor-origin cell type), Fibroblasts (key stromal components, though not prominently displayed in these specific plots), Macrophages (critical immune modulators), and T cells (CD4+ and CD8+). The goal is to identify how cellular communication networks change in PDAC, highlighting pathways relevant to disease progression and potential therapeutic intervention.

Visual Summary

Two dot plots are presented, illustrating significant cell-cell interactions for 'Adj_normal' and 'PDAC' conditions, respectively.

• The x-axis represents specific ligand-receptor pairs or complexes.
• The y-axis denotes interacting cell type pairs (e.g., "Mac|Diploid Ductal" where Macrophages are cell_A and Diploid Ductal cells are cell_B).
• Dot size reflects the statistical significance of the interaction, with larger dots indicating a smaller p-value (more significant).
• Dot color indicates the mean expression level of the ligand-receptor pair across the interacting cell populations, with warmer colors (yellow/green) signifying higher expression and cooler colors (purple/blue) indicating lower expression.

In the Adj_normal plot, interactions are predominantly observed between T CD8+ cells and Macrophages, and among T CD8+ cells themselves. Key interactions include HLA-E family members with various receptors, ICAM1-integrin complexes, and CD58-CD2. These interactions typically reflect immune surveillance and cell adhesion within healthy tissue. Notably, Ductal cells and Fibroblasts do not show prominent interactions within the top 80 pairs displayed for Adj_normal.

The PDAC plot reveals a markedly different landscape. Interactions become more numerous and involve Ductal cells (specifically 'Diploid Ductal' cells) extensively, communicating with Macrophages and other Ductal cells, alongside continued T cell and Macrophage interactions. Several high-expression, highly significant interactions emerge or become more prominent in PDAC, particularly those related to immune suppression, tumor progression, and stromal remodeling.

Biological Interpretation

The comparison between Adj_normal and PDAC conditions reveals a profound shift in the cellular communication network, reflecting the establishment of an immunosuppressive and pro-tumorigenic tumor microenvironment (TME) in PDAC.

1. Emergence of Ductal Cell-Mediated Communication in PDAC

A key observation is the extensive involvement of 'Diploid Ductal' cells in cell-cell interactions within the PDAC microenvironment, which were largely absent in the Adj_normal state. This highlights the active role of tumor-origin cells (or pre-malignant/diploid tumor cells) in shaping the TME. These Ductal cells engage in critical crosstalk with Macrophages, and exhibit self-interactions, indicating altered cellular behavior and network topology in cancer.

2. Upregulation of Immunosuppressive Pathways in PDAC

Several interactions highly prominent in PDAC point towards significant immune evasion mechanisms:

• TGFB1-TGFBR1: This interaction is highly significant and expressed in both Macrophage-to-Ductal cell and Ductal cell-to-Macrophage directions in PDAC. The Transforming Growth Factor-beta (TGF-β) pathway is a notorious driver of immune suppression, promoting T cell anergy, supporting regulatory T cell development, and inducing fibrosis in the pancreatic TME. PubMed search: TGF-beta pancreatic cancer immune suppression
• LGALS9-HAVCR2 (Galectin-9-TIM-3): This axis shows strong interaction from Diploid Ductal cells to Macrophages in PDAC. Galectin-9, when binding to T cell immunoglobulin and mucin domain-containing protein 3 (TIM-3) on immune cells, is a potent inducer of T cell exhaustion and apoptosis, thereby dampening anti-tumor immunity. GeneCards: LGALS9, GeneCards: HAVCR2
• FEBP1-LILRB4: Observed from Diploid Ductal cells to Macrophages. LILRB4 (Leukocyte Immunoglobulin-Like Receptor B4) is an inhibitory receptor often expressed on myeloid cells, and its engagement can promote an immunosuppressive phenotype in tumor-associated macrophages (TAMs).

3. Pathways Driving Tumor Progression and Stromal Remodeling

The PDAC microenvironment is characterized by a dense desmoplastic stroma, and several interactions support this:

• SPP1-integrin: Prominent from Macrophages to Diploid Ductal cells. Secreted phosphoprotein 1 (SPP1, also known as Osteopontin) is frequently overexpressed in PDAC and contributes to tumor progression, invasion, and metastasis by promoting cell survival, migration, and angiogenesis. It often signals through integrin receptors. GeneCards: SPP1
• PDGFB-PDGFRB: Significant interaction from Diploid Ductal cells to Macrophages. Platelet-Derived Growth Factor B (PDGFB) and its receptor (PDGFRB) signaling are crucial for fibroblast activation, angiogenesis, and the extensive stromal reaction observed in PDAC, further supporting tumor growth. GeneCards: PDGFB
• PLAUR-integrin: Observed from Macrophages to Diploid Ductal cells. The urokinase-type plasminogen activator receptor (PLAUR) is involved in extracellular matrix degradation and cell migration, facilitating tumor invasion and metastasis.
• TNFSF12-TNFRSF12 (TWEAK-Fn14): This interaction from Diploid Ductal cells to Macrophages is implicated in cell proliferation, inflammation, and angiogenesis, contributing to tumor growth and progression.

4. Immune Cell Cross-talk

While the Adj_normal condition showed robust interactions like HLA-E with NK cell receptors (e.g., NKG2A, KLRB1) on T CD8+ cells, these are less pronounced in the displayed top interactions in PDAC. This might suggest an altered or suppressed immune recognition mechanism in the tumor. ICAM1-integrin and CD58-CD2 interactions, crucial for cell adhesion and co-stimulation, remain active among T cells and macrophages in both conditions, indicating ongoing fundamental immune cell processes despite the pathological shifts.

Clinical or Translational Implications

The identified shifts in cell-cell communication networks in PDAC offer several critical insights for therapeutic development and biomarker discovery:

• Therapeutic Target Prioritization: The highly activated immunosuppressive pathways, such as TGFB1-TGFBR1 and LGALS9-HAVCR2 (Galectin-9-TIM-3), represent attractive therapeutic targets. Inhibitors of TGF-β signaling are already in clinical trials for various cancers, including PDAC, aimed at reversing immune suppression and reducing desmoplasia. Similarly, TIM-3 blockade is an emerging strategy to reinvigorate exhausted T cells.
• Targeting pro-tumorigenic and stromal remodeling pathways like SPP1-integrin and PDGFB-PDGFRB could also be beneficial. Disrupting SPP1 signaling could impede tumor invasion and metastasis, while inhibiting PDGF-R could reduce the dense desmoplastic stroma, potentially improving drug delivery and T cell infiltration.
• The TNFSF12-TNFRSF12 (TWEAK-Fn14) axis could also be explored, as its inhibition might reduce tumor growth and inflammation.
• Combinatorial Therapy Strategies: Given the complex interplay of these pathways, a single-agent approach may be insufficient. The data strongly support the rationale for combinatorial therapies that simultaneously target immune suppression (e.g., TIM-3 or TGF-β blockade) alongside strategies to disrupt the pro-tumorigenic stromal support (e.g., SPP1 or PDGF-R inhibition).
• Biomarker Development and Patient Stratification: The expression levels and activity of these specific ligand-receptor pairs (e.g., high TGFB1, SPP1, LGALS9 in the TME) could serve as prognostic biomarkers for PDAC patients or help stratify patients who are more likely to respond to therapies targeting these pathways. For instance, patients with high LGALS9-HAVCR2 interactions might benefit more from TIM-3 blocking antibodies.
• Experimental Validation: The insights gained from this analysis provide a strong basis for further experimental validation. *In vitro* co-culture experiments and *in vivo* mouse models of PDAC could be used to:
• Confirm the functional significance of these specific ligand-receptor interactions on tumor cell proliferation, migration, and immune cell function.
• Test the efficacy of blocking antibodies or small molecule inhibitors against the identified pathways (e.g., TGF-β inhibitors, TIM-3 blockers, SPP1 antagonists) alone or in combination.
• Investigate the impact of these interventions on the overall TME composition and immune cell infiltration.

11. Pancreatic Cell-Cell Interaction Analysis of Immune Checkpoint and Cell Cycle Pathways in Normal and PDAC Conditions

[Analysis Visualization Results]...

Analysis Overview

This analysis investigates cell-cell interactions (CCI) within pancreatic tissue, focusing on a curated set of genes associated with immune checkpoint and cell cycle pathways. The plot_cci_dots tool was used to visualize significant ligand-receptor interactions across various cell type pairs, comparing an adjacent normal pancreatic microenvironment (Adj_normal) with that of Pancreatic Ductal Adenocarcinoma (PDAC). The analysis specifically highlights interactions among cell types such as T cells (CD8+, CD4+), Macrophages, Endothelial cells, and Acinar cells, which are key components of the pancreatic tissue and tumor microenvironment. While the initial gene list included both immune checkpoint and cell cycle regulators, the CCI analysis predominantly reveals ligand-receptor interactions, naturally emphasizing the immune checkpoint-related components.

Visual Summary

CCI for Adj_normal

The plot for Adj_normal pancreas displays a diverse and intricate network of cell-cell interactions involving the selected gene set.

• Dominant Pathways: Interactions related to the TGF-beta family (e.g., TGFB1_TGFbeta_receptor1, TGFB1_integrin_avb6_complex, TGFB3_TGFbeta_receptor1, TGFB3_integrin_avb6_complex) are highly prominent, especially within Acinar cells (Acinar|Acinar) and between Acinar cells and Macrophages (Acinar|Mac) or Endothelial cells (Endo|Acinar). These interactions show consistently high mean expression (yellow/green dots) and strong statistical significance (large dots).
• EGFR Signaling: EGF_EGFR and HBEGF_EGFR interactions are also notable, particularly involving Endothelial|Acinar and Mac|Acinar cell pairs, suggesting active epithelial growth factor signaling.
• Immune Cell Interactions: T CD8+ cells exhibit self-interactions (T CD8+|T CD8+) via LCK_CD8_receptor (a proxy for CD8-mediated T cell signaling) and interactions with Macrophages, Endothelial cells, and Acinar cells, often involving IFNGR1 (IFNG_Type_II_IFNR) and CD93_IFNGR1. This indicates active immune surveillance and communication in the normal pancreatic microenvironment.

CCI for PDAC

In contrast, the PDAC plot shows a markedly different and less diverse pattern of cell-cell interactions for the same gene set.

• Reduced Diversity: Fewer significant interactions are displayed compared to the Adj_normal condition, suggesting a remodeling or restriction of prominent communication pathways among the selected ligand-receptor pairs.
• Immune Cell-Centric Interactions: The majority of interactions observed in PDAC are concentrated among immune cells, specifically T cells (T CD8+, T CD4+) and Macrophages.

Prominent Co-stimulation and IFN-gamma Signaling

• The CD86_CD28 interaction is notably strong and significant, observed in Macrophage|T CD4+ and T CD8+|T CD8+ pairs. CD86 on antigen-presenting cells (like Macrophages) provides a crucial co-stimulatory signal to CD28 on T cells.
• IFNG_Type_II_IFNR (IFNGR1) interactions are present in T CD4+|T CD8+ and Mac|Mac pairs, indicating ongoing IFN-gamma signaling within the immune compartment.
• LCK_CD8_receptor interactions persist in T CD8+|T CD8+ and T CD8+|Mac pairs.
• Loss of Acinar-Related and TGF-beta/EGFR Prominence: Interactions involving Acinar cells, which were prominent in Adj_normal, are largely absent or significantly diminished for these specific gene pairs in the PDAC context. Similarly, the strong TGF-beta and EGFR signaling observed in Adj_normal are not prominently featured in the PDAC plot for the selected ligand-receptor pairs, with TGFB1_TGFbeta_receptor1 only showing a very weak interaction in one pair.

Biological Interpretation

1. Homeostatic Maintenance in Normal Pancreas: The robust TGF-beta and EGFR signaling observed in Adj_normal pancreatic tissue underscores their critical roles in maintaining tissue homeostasis, regulating acinar cell proliferation, differentiation, and influencing local immune responses. TGF-beta is a known regulator of cell growth and differentiation, often playing a role in immune suppression and fibrosis in various tissues PubMed Search: TGF-beta pancreas homeostasis.
2. Remodeling of the PDAC Microenvironment: The stark contrast in CCI patterns between Adj_normal and PDAC conditions signifies a profound remodeling of intercellular communication within the tumor microenvironment. The reduced diversity and shift towards immune cell-centric interactions suggest a dysregulated network that may contribute to tumor progression and immune evasion.
3. Dysfunctional Immune Activation in PDAC: The prominence of the CD86-CD28 co-stimulatory axis between Macrophages and T CD4+ cells, and within T CD8+ cells, in PDAC is a key finding. While CD28 co-stimulation is essential for T cell activation, in the context of PDAC, this activation often fails to mount an effective anti-tumor response and can even contribute to T cell exhaustion or the generation of regulatory T cells, promoting immune evasion GeneCards: CD28. Similarly, IFN-gamma signaling is critical for anti-tumor immunity, but sustained or dysregulated IFN-gamma signaling in the tumor microenvironment can also induce PD-L1 expression on tumor cells, leading to T cell anergy and exhaustion PubMed Search: IFN-gamma PD-L1 tumor microenvironment.
4. Altered Role of Acinar Cells and Associated Signaling: The diminished representation of Acinar cell interactions and the specific TGF-beta/EGFR ligand-receptor pairs in PDAC suggests that these pathways, while critical in normal physiology, may be hijacked or superseded by other mechanisms in the tumor. Given that Ductal cells are the tumor origin, the focus shifts away from Acinar cell-mediated communication for these specific cancer-relevant gene sets.

Clinical or Translational Implications

1. Therapeutic Target Prioritization in PDAC Immunotherapy: The significant CD86-CD28 and IFN-gamma interactions within the immune cell compartment in PDAC highlight critical axes for potential therapeutic intervention. Modulating the CD86-CD28 pathway could involve strategies to enhance effective T cell activation or to block inhibitory signals if it contributes to immunosuppression. Targeting IFN-gamma signaling might involve fine-tuning its effects to prevent immune evasion while preserving beneficial anti-tumor responses.
2. Understanding Immune Evasion Mechanisms: The observed patterns contribute to understanding how PDAC creates an immunosuppressive microenvironment despite ongoing immune cell communication. Further investigation into the functional consequences of these specific interactions (e.g., whether CD28 signaling leads to effector T cell activation or Treg expansion) is crucial.
3. Biomarker Discovery: The identified specific cell-cell interaction patterns (e.g., strong Mac|T CD4+ CD86-CD28 interaction in PDAC) could serve as potential biomarkers for distinguishing PDAC from normal tissue, predicting disease progression, or assessing response to immunotherapies.
4. Experimental Validation: The specific ligand-receptor pairs identified as significant in PDAC (e.g., CD86-CD28, IFNG-IFNGR1) warrant further experimental validation using *in vitro* co-culture models, organoids, or *in vivo* animal models to confirm their functional roles in PDAC development and immune evasion. This could pave the way for novel therapeutic strategies.

12. Condition-Specific Cell-Cell Interaction Patterns in PDAC

[Analysis Visualization Results]...

Analysis Overview

This analysis identifies cell-cell interactions (CCI) that significantly differ between Pancreatic Ductal Adenocarcinoma (PDAC) and adjacent normal pancreatic tissue (Adj_normal). The focus is on interactions involving major immune cells (T cells, Myeloid cells/Macrophages, Mast cells, B cells) and stromal cells. The plot_dot_for_cci_with_signif_difference tool was used to visualize these differences, displaying interaction strength and significance for individual samples within each condition.

Visual Summary

The dot plot effectively illustrates the condition-specific cell-cell interaction landscape:

• Strong Differential Activity: A striking difference is observed between the two conditions. Samples from Adj_normal tissue generally exhibit very few and weak cell-cell interactions (lighter red, smaller dots). In stark contrast, PDAC samples display a widespread and robust pattern of strong and highly significant interactions (darker red, larger dots) across a multitude of ligand-receptor pairs.
• Dominant Interaction Types: The most prominent interactions involve Macrophages (Mac) and CD8+ T cells (T CD8+), with many CCI indices representing Mac|T CD8+ or T CD8+|Mac pairs. Some interactions also involve NK cells (Mac|NK) and Mast cells (T CD8+|Mast cell).
• Specific CCI Hotspots in PDAC: Numerous interactions are highly activated almost exclusively in PDAC samples. Notable examples include:
• Immune regulatory axes: SIRPA-CD47 (Mac|T CD8+), HLA-E-CD94:NKG2A (Mac|T CD8+), FASLG-FAS (T CD8+|Mac), TNFRSF14-TNFSF14 (T CD8+|Mac).
• Adhesion and migration molecules: ICAM1-integrin complexes (Mac|Mac, Mac|T CD8+).
• Signaling molecules: SEMA4A-PLXND1 (Mac|Mac), SEMA4D-PTPRC (Mac|Mac), CCL4-CCR5 (T CD8+|IT CD8+).
• Quantifiable Differences: The color intensity (Scaled log strength) clearly shows higher interaction strengths in PDAC, while larger dot sizes (-log10(p)) indicate greater statistical significance for these interactions in the tumor microenvironment compared to adjacent normal tissue.

Biological Interpretation

The observed patterns highlight a dramatic remodeling of the immune and stromal cell interaction network within the PDAC tumor microenvironment (TME), which is largely quiescent in adjacent normal tissue.

1. Macrophage-T Cell Crosstalk as a Central Feature of PDAC TME: The overwhelming presence of macrophage-T cell CD8+ interactions underscores their critical role in PDAC biology. Macrophages, particularly tumor-associated macrophages (TAMs), are abundant in PDAC and are known to contribute to immunosuppression, tumor progression, and metastasis [PubMed Search]. The strong activation of these interactions suggests a heightened and dysregulated dialogue between these key immune cell populations in the tumor.
2. Immune Evasion and Suppression Mechanisms:

• SIRPA-CD47 (Mac|T CD8+): CD47, often referred to as a "don't eat me" signal, interacts with SIRPA on phagocytic cells. Its upregulation in PDAC is a known mechanism for tumor cells to evade macrophage-mediated phagocytosis [PubMed Search]. While typically discussed in the context of cancer cells, its strong activation between Macrophages and CD8+ T cells here suggests complex regulatory roles within the immune compartment itself, possibly modulating T cell responses or macrophage-T cell interactions.
• HLA-E-CD94:NKG2A (Mac|T CD8+): HLA-E, an MHC class I molecule, can present self-peptides and interact with the inhibitory receptor NKG2A found on NK cells and certain T cells. This interaction can suppress cytotoxic functions, representing a crucial immune evasion strategy employed by tumors to protect against T cell and NK cell attack [PubMed Search].
• FASLG-FAS (T CD8+|Mac): The Fas ligand (FASLG) and its receptor Fas (FAS) axis is a major pathway for inducing apoptosis. Upregulated FASLG in the TME can trigger apoptosis in activated T cells, leading to T cell exhaustion and immune tolerance [PubMed Search]. Its strong presence here points to a mechanism of T cell suppression in PDAC.

3. Cell Adhesion and Migration: Interactions involving ICAM1 and integrin complexes are vital for immune cell trafficking, adhesion, and antigen presentation. Their heightened activity in PDAC samples may reflect increased immune cell infiltration into the TME or altered adhesive properties of these cells within the desmoplastic stroma characteristic of PDAC.
4. Role of Semaphorins: Semaphorins (e.g., SEMA4A, SEMA4D) are axon guidance molecules increasingly recognized for their diverse roles in immune regulation, including modulating T cell activation, differentiation, and macrophage polarization [PubMed Search]. Their enhanced interactions between macrophages and T cells suggest their involvement in shaping the immune landscape of PDAC.
5. Mast Cell and NK Cell Contributions: While less numerous than macrophage-T cell interactions, the presence of active interactions involving NK cells (e.g., SELL-NK|Mac) and Mast cells (e.g., PTPRC_CD22-T CD8+|Mast cell) indicates their participation in the altered TME. Mast cells can have both pro- and anti-tumorigenic roles, while NK cells are crucial for innate immunity against cancer [PubMed Search].

Clinical or Translational Implications

The distinct and robust cell-cell interaction patterns observed in PDAC, particularly those involving macrophages and T cells, offer several translational avenues:

• Biomarker Discovery: The specific ligand-receptor pairs that are significantly enhanced in PDAC could serve as novel biomarkers for early detection, diagnosis, or prognosis of PDAC. Monitoring the activity of interactions like SIRPA-CD47 or HLA-E-NKG2A could provide insights into disease progression or response to therapy.
• Therapeutic Targeting: Many of the highly active interaction axes identified (e.g., SIRPA-CD47, HLA-E-NKG2A, FASLG-FAS) represent established or emerging targets in cancer immunotherapy. Modulating these interactions, for instance, by blocking inhibitory pathways or enhancing pro-apoptotic signals, could reprogram the PDAC TME to favor anti-tumor immunity.
• Combination Immunotherapy Strategies: Given the complex and often immunosuppressive nature of the PDAC TME, understanding these specific intercellular dialogues is crucial for developing rational combination immunotherapies. For example, combining anti-CD47 therapies with T cell-focused immunotherapies could potentially overcome existing resistance mechanisms in PDAC.
• Understanding Treatment Resistance: These findings can help explain mechanisms of resistance to current immunotherapies in PDAC. The active immunosuppressive CCIs identified might be key players in dampening the efficacy of T-cell-directed therapies, highlighting the need to address the broader TME.

13. Ductal Cell Condition-Specific Surfaceome Markers in Pancreatic Ductal Adenocarcinoma (PDAC)

[Analysis Visualization Results]...

Analysis Overview

This analysis identifies condition-specific surfaceome markers for Ductal cells, comparing normal adjacent pancreas tissue (Adj_normal) with Pancreatic Ductal Adenocarcinoma (PDAC) tissue. The results are visualized as a dot plot, where each row represents a sample and each column represents a gene. The color intensity of the dot indicates the mean expression level of the gene within that sample's Ductal cells, and the size of the dot represents the fraction of Ductal cells in that sample expressing the gene. The analysis specifically focused on surfaceome markers, identifying at most 50 markers per condition, making them particularly relevant for cell-surface-targeted interventions or diagnostics. Ductal cells are noted as the tumor origin cell type. Samples are grouped by condition and ploidy status (Diploid PDAC vs. likely Aneuploid PDAC, simply labeled "PDAC" in the plot).

Visual Summary

The dot plot effectively highlights distinct expression patterns of surfaceome markers across different conditions and sample types of Ductal cells:

• Adj_normal Ductal Cells: A small group of markers, including GP2 and GPRC5A, show relatively higher expression and prevalence in the Adj_normal samples (AdjN_1, AdjN_2, AdjN_3) compared to all PDAC groups. However, the overall expression levels and cell fractions for these markers are not as uniformly high as those observed in the prominent PDAC tumor cell clusters.
• Diploid PDAC Ductal Cells: Ductal cells from PDAC samples identified as "Diploid PDAC" (e.g., Diploid PDAC_11B, Diploid PDAC_4) exhibit a mixed expression pattern. They generally show very low to absent expression for most markers highly expressed in the "PDAC" group (red box), but some markers like TSPAN1, PLAUR, VSIG2, QSOX1, ITGA2, ERBB3, MET, CDCP1, SLC2A1, and EMP1 show intermediate expression levels or higher cell fractions compared to Adj_normal, albeit lower than the main PDAC tumor cells. This suggests an altered state compared to normal, potentially reflecting early molecular changes or interaction with the tumor microenvironment.
• PDAC Tumor Ductal Cells (Likely Aneuploid): The group labeled "PDAC" (highlighted by the red box, e.g., PDAC_15, PDAC_1, PDAC_6) shows a dramatically upregulated and widespread expression of numerous surfaceome markers. These samples likely represent the transformed, aneuploid tumor cells, consistent with Ductal cells being the tumor origin. Key markers highly expressed in this group, often across a high fraction of cells and with strong intensity, include:
• QSOX1, TSPAN1, TM9SF2, PLAUR, PDLIM5, ITGA2, PRSS8, ERBB3, NPTN, CXADR, SLC44A4, MET, CDCP1, CXCL16, TSPAN15, LDLR, MYOF, AMN, SLC2A1, CLSTN1, SCNN1A, MYADM, CRB3, UNC93B1, EFNA1, PTPRA, FABP1, ERBB2, EMP1, F3, CLDN12, PCDH1, SUCO, SLC52A2, SEZ6L2, MPZL1, TMEM87B, ATP2B4, ANTXR2, ZDHHC5, LTBR, PAM, ADAM15, CEACAM1, DAG1, EFNA5.
• Many of these markers are expressed across nearly all cells in these tumor samples (large dot size) and at high levels (dark red color).

Biological Interpretation

The distinct expression profiles of surfaceome markers in Ductal cells reflect profound biological changes associated with PDAC development and progression.

• Normal Ductal Cell Markers: The limited specific markers for Adj_normal Ductal cells, such as GP2 (a zymogen granule membrane protein) and GPRC5A (a G-protein coupled receptor), suggest their roles in normal pancreatic function or potential protective mechanisms that are lost or downregulated in cancer. GPRC5A can exhibit tumor-suppressive functions in certain contexts [1].
• Emerging Malignancy in Diploid PDAC Cells: The "Diploid PDAC" group, while not as starkly altered as the main PDAC tumor cells, shows intermediate expression of some genes typically associated with cancer (e.g., TSPAN1, PLAUR, MET, SLC2A1). This might indicate that even diploid ductal cells within the tumor microenvironment are undergoing early transcriptional changes or receiving signals that predispose them to malignant transformation, prior to significant aneuploidy. These cells could represent a transitional state or reactive ductal cells.
• Hallmarks of Cancer in Aneuploid PDAC Cells: The widespread and high expression of a large panel of surfaceome markers in the "PDAC" (likely aneuploid tumor) group strongly points to fundamental alterations driving pancreatic carcinogenesis. Many of these genes are well-established oncogenes or involved in critical cancer processes:
• Cell Proliferation and Survival: ERBB2 (HER2) and ERBB3 (HER3) are receptor tyrosine kinases frequently dysregulated in cancers, promoting cell growth and survival pathways [2]. MET (HGFR) is another receptor tyrosine kinase involved in cell proliferation, survival, invasion, and angiogenesis, often activated in PDAC [3].
• Metabolic Reprogramming: SLC2A1 (GLUT1), a glucose transporter, is a classic marker of the Warburg effect, where cancer cells rely on aerobic glycolysis for energy, fueling rapid growth [4].
• Cell Adhesion, Migration, and Invasion: TSPAN1, PLAUR (uPAR), ITGA2 (Integrin Alpha 2), CDCP1, CEACAM1, CLDN12, PCDH1 are all involved in cell adhesion, extracellular matrix interaction, and invasion, crucial for tumor spread and metastasis [5, 6, 7]. For example, uPAR (PLAUR) plays a central role in pericellular proteolysis, facilitating cancer cell invasion.
• Tumor Microenvironment Interaction: CXCL16 is a chemokine that can influence immune cell recruitment, potentially modulating the immune landscape within the tumor.
• Enzymatic Activity and Secretion: QSOX1 and PRSS8 (Prostasin) are enzymes whose dysregulation contributes to the proteolytic environment favoring tumor invasion and growth.

The "surfaceome only" constraint means these markers are located on the cell surface, making them readily accessible for various applications.

Clinical or Translational Implications

The identification of these condition-specific surfaceome markers, particularly those highly upregulated in PDAC tumor ductal cells, has significant clinical and translational implications:

1. Diagnostic Biomarkers: Genes like QSOX1, TSPAN1, PLAUR, MET, SLC2A1, ERBB3, CDCP1, CEACAM1, which show high expression and prevalence in PDAC tumor cells but low expression in normal or diploid PDAC cells, could serve as highly specific diagnostic biomarkers for PDAC. These could be detected in tissue biopsies, liquid biopsies (e.g., circulating tumor cells, exosomes), or through imaging techniques to identify tumor cells.
2. Therapeutic Targets: As surfaceome proteins, many of these markers are excellent candidates for targeted therapies.

• Antibody-drug conjugates (ADCs): Targeting highly expressed surface proteins like ERBB2, ERBB3, MET, CDCP1, TSPAN1, PLAUR, CEACAM1 with ADCs could deliver cytotoxic drugs specifically to tumor cells while sparing normal tissues.
• Immunotherapy: Monoclonal antibodies blocking the function of pro-tumorigenic receptors (e.g., MET, ERBB3) or modulating immune interactions (e.g., via CXCL16) could be developed.
• Small molecule inhibitors: For receptors like MET and ERBB3, existing or novel small molecule inhibitors could be explored in PDAC.

3. Prognostic Markers: The expression levels of certain markers could correlate with disease aggressiveness, stage, or patient outcomes, aiding in prognosis and treatment stratification.
4. Early Detection and Monitoring: Markers specifically expressed in "Diploid PDAC" cells but not normal cells might indicate early disease states or pre-malignant lesions, offering opportunities for early detection and intervention.
5. Experimental Validation: These identified markers warrant further experimental validation using techniques such as immunohistochemistry, immunofluorescence, flow cytometry, or functional assays in PDAC cell lines and patient-derived organoids to confirm their specificity and functional roles.

The clear distinction between normal and malignant ductal cell surfaceomes provides a rich resource for developing novel strategies against PDAC, a highly aggressive and challenging cancer.

---

References:

[1] GPRC5A: A Versatile Protein in Human Diseases. PubMed search: https://pubmed.ncbi.nlm.nih.gov/?term=GPRC5A+cancer

[2] HER3 (ERBB3): A Key Player in Tumorigenesis and a Promising Therapeutic Target. GeneCards: https://www.genecards.org/cgi-bin/carddisp.pl?gene=ERBB3

[3] MET (MET proto-oncogene): GeneCards: https://www.genecards.org/cgi-bin/carddisp.pl?gene=MET

[4] SLC2A1 (GLUT1): GeneCards: https://www.genecards.org/cgi-bin/carddisp.pl?gene=SLC2A1

[5] PLAUR (uPAR): GeneCards: https://www.genecards.org/cgi-bin/carddisp.pl?gene=PLAUR

[6] CDCP1: GeneCards: https://www.genecards.org/cgi-bin/carddisp.pl?gene=CDCP1

[7] CEACAM1: GeneCards: https://www.genecards.org/cgi-bin/carddisp.pl?gene=CEACAM1

14. Macrophage Condition-Specific Surfaceome Markers in Pancreatic Cancer

[Analysis Visualization Results]...

Analysis Overview

This analysis aimed to identify surfaceome markers that are differentially expressed in Macrophage cells between adjacent normal pancreas (Adj_normal) and pancreatic ductal adenocarcinoma (PDAC) conditions. The plot_markers_and_expression_dot tool was used to visualize the expression patterns of these markers across individual samples. The selection was specifically filtered to include only surfaceome markers, making them strong candidates for targeted interventions.

Visual Summary

The dot plot displays the mean expression level (color intensity) and the fraction of cells expressing each marker (dot size) for Macrophage cells across different samples, categorized by their condition (Adj_normal or PDAC).

1. Distinct Expression Profiles: There is a striking difference in surfaceome marker expression between Adj_normal and PDAC samples. Macrophages from Adj_normal samples (AdjN_2, AdjN_3, AdjN_1) consistently show very low expression (light or white dots) and low prevalence (small dots) for almost all markers.
2. PDAC-Associated Upregulation: In contrast, Macrophages from PDAC samples exhibit robust and widespread upregulation of a multitude of surfaceome markers. Many genes, such as GP2, ITGAX, SIRPA, BSG, IGF2R, HLA-F, TNFRSF14, PLXNC1, IL6R, TGFBR2, ADAM8, ADAM10, LILRB2, CSF2RB, CD300LF, and LY6E, show high mean expression (dark red dots) and high prevalence (large dots) across multiple PDAC samples.
3. Heterogeneity within PDAC: While most PDAC samples show strong upregulation, there is some variability. Samples like PDAC_8, PDAC_5, PDAC_15, PDAC_3, PDAC_7, PDAC_16, PDAC_12, and PDAC_2 show particularly strong and broad expression of most markers. Other PDAC samples (e.g., PDAC_9, PDAC_1, PDAC_6, PDAC_10, PDAC_4, PDAC_13, PDAC_11B, PDAC_11A) also show upregulation but with more heterogeneous patterns or lower average expression for some specific markers. This suggests potential phenotypic diversity among macrophages even within the PDAC microenvironment across different patients.
4. Cell Counts: The bar plot on the right indicates the number of Macrophage cells identified in each sample. The varying cell counts (e.g., 24 cells in PDAC_16 vs. 2910 cells in PDAC_9) should be considered, though the overall trend of PDAC-specific marker enrichment is consistent across samples with sufficient cell numbers.

Biological Interpretation

The observed upregulation of numerous surfaceome markers in Macrophages from PDAC samples points towards a distinct, tumor-associated macrophage (TAM) phenotype that differs significantly from macrophages in the adjacent normal pancreatic tissue. This is consistent with the established role of TAMs in promoting tumor progression in PDAC.

Key biological insights from the identified markers include:

• Immune Modulation and Suppression: Several upregulated markers are known to play roles in immune regulation.
• SIRPA (CD172a) is an inhibitory receptor that interacts with CD47 on tumor cells, leading to "don't eat me" signals that protect cancer cells from phagocytosis. Its upregulation suggests an immune suppressive role for TAMs in PDAC. UniProt: SIRPA
• LILRB2 (ILT4, CD85d) is another immune checkpoint receptor that can suppress immune responses, often associated with myeloid-derived suppressor cells (MDSCs) and TAMs in cancer. GeneCards: LILRB2
• CD46 is a complement regulatory protein expressed on many cell types, including macrophages, potentially protecting them from complement-mediated lysis within the TME.
• TNFRSF14 (HVEM) can act as a co-stimulatory or co-inhibitory receptor depending on its ligand, and its expression on TAMs can contribute to the complex immune landscape of the tumor.
• HLA-F, a non-classical MHC class I molecule, can be involved in immune evasion by tumor cells and immune cells alike.

Cell Adhesion and Extracellular Matrix (ECM) Remodeling:

• ITGAX (CD11c) is an integrin involved in cell adhesion and migration, indicating active trafficking and interaction of TAMs with the tumor microenvironment. GeneCards: ITGAX
• BSG (CD147, Basigin) promotes matrix metalloproteinase production and plays a role in cell invasion and angiogenesis. UniProt: BSG
• PLXNC1 (Plexin C1) is a receptor involved in cell guidance and immune cell migration.
• ADAM8 and ADAM10 are metalloproteases that can shed cell surface proteins and cleave ECM components, contributing to tumor invasion and metastasis. GeneCards: ADAM10

Growth Factor and Cytokine Signaling:

• IL6R (CD126) is the receptor for IL-6, a pro-inflammatory cytokine highly abundant in the PDAC TME, promoting tumor growth, metastasis, and immune evasion. Upregulation of IL6R suggests TAMs are highly responsive to IL-6 signaling. GeneCards: IL6R
• TGFBR2 is a receptor for TGF-β, a potent immunosuppressive and pro-fibrotic cytokine in PDAC. Increased TGFBR2 on TAMs suggests enhanced responsiveness to TGF-β, contributing to their pro-tumorigenic phenotype. UniProt: TGFBR2
• IGF2R (M6P/IGF2R) can mediate various cellular functions, including growth regulation.

Other Noteworthy Markers:

• GP2 (Glycoprotein 2) can be associated with specific immune cell subsets and inflammatory processes.
• LY6E is frequently expressed in aggressive cancer cells and cancer stem cells, and its upregulation on TAMs might indicate a more aggressive or activated phenotype. GeneCards: LY6E

The broad and consistent upregulation of these markers in PDAC macrophages indicates a significant reprogramming of these immune cells within the tumor microenvironment, contributing to an immunosuppressive and pro-tumorigenic milieu. The observed heterogeneity across PDAC samples highlights the complexity of TAM populations, suggesting diverse functional states or origins.

Clinical or Translational Implications

The identification of these highly expressed, condition-specific surfaceome markers on Macrophages in PDAC offers several clinical and translational opportunities:

1. Biomarker Development: The consistently upregulated markers, particularly those with high mean expression and prevalence across multiple PDAC samples, could serve as potential diagnostic or prognostic biomarkers for PDAC. Their presence on macrophages could indicate tumor burden, disease progression, or response to therapy.
2. Therapeutic Targets: Given that these are surfaceome markers, they are highly accessible for targeted therapies.

• Antibody-drug conjugates (ADCs) or CAR (Chimeric Antigen Receptor) therapies (e.g., CAR-macrophages or CAR-T cells) could be engineered to specifically target and eliminate or reprogram these pro-tumorigenic macrophages in PDAC.
• Targeting receptors like IL6R or TGFBR2 could directly inhibit crucial pro-tumorigenic signaling pathways that are active within the TAMs.
• Targeting immune checkpoints such as SIRPA or LILRB2 could reverse the immune suppressive functions of TAMs, thereby enhancing anti-tumor immunity. PubMed search: TAM targeting therapy PDAC

3. Patient Stratification and Personalized Medicine: The heterogeneity observed in marker expression across PDAC samples suggests that a single therapeutic approach might not be effective for all patients. Stratifying patients based on the specific surfaceome marker profiles of their tumor macrophages could enable personalized treatment strategies.
4. Experimental Validation: These findings warrant further experimental validation to confirm protein expression on the cell surface using techniques such as flow cytometry, immunohistochemistry, or spatial proteomics. Functional studies would also be crucial to elucidate the precise roles of these markers in TAM biology and PDAC progression, paving the way for preclinical and clinical development.

15. Comprehensive Surfaceome Marker Expression for Pancreatic Cell Subsets

[Analysis Visualization Results]...

Analysis Overview

This analysis aimed to identify and visualize surfaceome markers for various celltype_subset populations within the human pancreas single-cell RNA-seq dataset. While the initial query specified "condition-specific markers for Fibroblast," the applied tool parameters (target_cell=None) generated a broader dot plot displaying cell-type specific surfaceome markers across *all* annotated cell subsets. This visualization is crucial for validating the assigned cell type identities and understanding their molecular profiles based on surface proteins.

Visual Summary

The provided dot plot effectively summarizes the expression patterns of surfaceome markers across 35 distinct celltype_subset populations.

• Axes: The Y-axis lists the celltype_subset categories, and the X-axis displays the identified surfaceome marker genes.
• Dot Size: Represents the fraction of cells within each group (cell type) that express a given gene. Larger dots indicate higher prevalence of expression.
• Dot Color Intensity: Represents the mean expression level of the gene within the expressing cells of that group. Darker red indicates higher mean expression.
• Diagonal Specificity: A prominent pattern is the presence of red boxes along the diagonal, highlighting clusters of genes that are highly expressed and specific to their respective celltype_subset. This diagonal pattern indicates strong specificity of the selected markers for their annotated cell types, which is ideal for cell identity confirmation.
• Fibroblast Markers: For Fibroblast cells, a distinct set of highly expressed and specific surfaceome markers is observed, including genes like DCN, LUM, COL1A1, COL3A1, COL6A2, PDGFRA, FAP, and ACTA2. Similar specific marker sets are visible for other cell types.
• Minority Populations: Some cell types, especially those with fewer cells (indicated by the bar chart on the right), still exhibit clear marker specificity, demonstrating the robustness of the marker identification.

Biological Interpretation

The observed patterns of surfaceome marker expression strongly support the celltype_subset annotations and provide valuable insights into the cellular composition of the pancreas.

Fibroblast Identity Confirmation:

• The Fibroblast population is clearly delineated by markers such as DCN (Decorin) and LUM (Lumican), both small leucine-rich proteoglycans involved in extracellular matrix (ECM) organization and collagen fibrillogenesis UniProt: P21810, UniProt: P07585.
• Collagen genes like COL1A1, COL3A1, and COL6A2 are highly expressed, consistent with their role in producing and remodeling the ECM.
• PDGFRA (Platelet-Derived Growth Factor Receptor Alpha) is a well-known fibroblast marker, involved in cell growth, proliferation, and differentiation GeneCards: PDGFRA.
• FAP (Fibroblast Activation Protein) is a highly specific marker for activated fibroblasts, particularly prominent in tumor-associated fibroblasts in pancreatic ductal adenocarcinoma (PDAC) PubMed search: FAP pancreatic cancer.
• ACTA2 (Actin Alpha 2, Smooth Muscle) indicates the presence of myofibroblast-like cells, which are activated fibroblasts characterized by contractile properties and high ECM production, often found in fibrosis and tumor stroma.

Distinct Cell Type Annotations:

• Acinar cells are robustly identified by pancreatic digestive enzyme genes like PRSS1 (Trypsin-1), CPA1 (Carboxypeptidase A1), and CELA3A (Chymotrypsin-like Elastase 3A), reflecting their exocrine function.
• Ductal cells show strong expression of KRT7 (Keratin 7), KRT19 (Keratin 19), and CLDN4 (Claudin 4), which are canonical epithelial markers characteristic of pancreatic ducts.
• B cell subsets (Breg, Follicular, MZ, Memory, Plasma) are identifiable by pan-B cell markers such as CD79A, CD79B, CD19, and CD22, with Plasma cells additionally expressing MZB1 and XBP1 related to antibody production.
• T cell subsets are distinguished by pan-T cell markers like CD3E, with subsets showing specific co-expression (e.g., CD4 for helper T cells, CD8A/B for cytotoxic T cells, FOXP3 for regulatory T cells (Treg)).
• Macrophage subsets (M1, M2A, M2B, M2C, M2D) express common macrophage markers like CD68 and CD163 (for M2-like macrophages), consistent with their roles in immune surveillance and tissue remodeling.
• Endothelial cells (including Endothelial tip cell and Lymphatic Endothelial cell) express genes like CDH5 (VE-cadherin), ENG (Endoglin/CD105), and FLT1 (VEGFR1), which are crucial for vascular integrity and angiogenesis. Lymphatic Endothelial cells are uniquely marked by PROX1.

Biological Nuances and Overlaps:

• The co-expression of genes like ACTA2, COL1A1, PDGFRA, and FAP in both Fibroblast and Stellate cell populations reflects the close functional relationship, particularly the activation of pancreatic stellate cells into myofibroblast-like cells in disease contexts like PDAC and fibrosis PubMed search: pancreatic stellate cell myofibroblast.
• The use of surfaceome markers is particularly insightful as these proteins are accessible on the cell surface, making them potential targets for cell sorting (e.g., FACS), imaging, and targeted therapies.

Annotation Notes

The comprehensive display of surfaceome markers provides strong evidence supporting the quality and distinctness of the celltype_subset annotations within this single-cell dataset. The clear diagonal specificity observed in the dot plot indicates that the current cell type assignments are well-supported by highly specific and differentially expressed surface proteins. This robust marker validation enhances confidence in downstream analyses that rely on these cell type classifications. The identification of established markers for diverse cell types, including various immune cell subtypes, stromal cells, and epithelial lineages, confirms the successful partitioning of the complex pancreatic cellular landscape.

16. T cell CD4+ Condition-Specific Surfaceome Markers in Pancreatic Cancer

[Analysis Visualization Results]...

Analysis Overview

This analysis aimed to identify surfaceome markers that are specifically enriched in CD4+ T cells from either "Adj_normal" (adjacent normal pancreas tissue) or "PDAC" (Pancreatic Ductal Adenocarcinoma) conditions. The plot_markers_and_expression_dot tool was employed, focusing exclusively on surface-expressed proteins. This focus is critical for identifying potential diagnostic biomarkers or therapeutic targets that are accessible on the cell surface. The resulting dot plot illustrates the mean expression level (color intensity) and the fraction of cells expressing each marker (dot size) across individual samples, which are grouped by their respective conditions.

Visual Summary

The dot plot effectively delineates distinct sets of surfaceome markers that characterize CD4+ T cells in either the "Adj_normal" or "PDAC" pancreatic tissue environments.

• Adj_normal-associated markers: A clear cluster of genes, including *SLC2A3, PTGER4, GPR171, GP2, AREG, IFNGR1, IL18R1, TNFSF8, CCR7, CLEC2D*, and *SELL*, exhibits higher mean expression and is present in a larger fraction of CD4+ T cells within the "Adj_normal" samples (AdjN_1, AdjN_2). Notably, *CCR7* and *SELL* show particularly strong and widespread expression in these normal samples.
• PDAC-associated markers: In contrast, a prominent group of markers, such as *BTN3A2, IL10RA, TNFRSF25, SUSD3, SPN, TM9SF3, SERINC3, CD28, SSR1, ICAM2, IFNAR2, GPR65, NUP210, TMEM106B, ATP2B4, CD82, ITGB7, TMEM63A*, and *IL6R*, is distinctly upregulated in most PDAC samples. For many of these, expression is observed across a large fraction of cells and at high mean levels (large, dark red dots).
• Sample Heterogeneity within PDAC: While many PDAC-associated markers show broad elevation, there is considerable heterogeneity in expression patterns among individual PDAC samples. Some samples (e.g., PDAC_16, PDAC_11A, PDAC_2) consistently exhibit lower expression (smaller, lighter dots) for several PDAC-enriched markers, suggesting variations in the tumor immune microenvironment or T cell states across different patients.
• Cell Counts per Sample: The bar chart to the right provides the total number of CD4+ T cells analyzed per sample. This context is important for interpreting the robustness of marker expression; samples with higher cell counts generally yield more reliable average expression estimates.

Biological Interpretation

The differential expression of these surfaceome markers on CD4+ T cells offers insights into the distinct functional states and adaptations of these immune cells within the healthy pancreas versus the immunosuppressive tumor microenvironment of PDAC.

• CD4+ T Cell Phenotype in Adj_normal Tissue: The strong upregulation of CCR7 and SELL (L-selectin) in Adj_normal CD4+ T cells is highly indicative of naive or central memory T cell populations https://pubmed.ncbi.nlm.nih.gov/11729013/. These markers facilitate homing to secondary lymphoid organs, suggesting a more recirculating or quiescent phenotype typical of immune surveillance in healthy tissue. The presence of IFNGR1 and IL18R1 suggests these cells maintain a capacity for robust inflammatory responses.
• CD4+ T Cell Adaptation in the PDAC Microenvironment: The markers enriched in PDAC CD4+ T cells point towards a T cell population that is engaged with, and likely shaped by, the tumor microenvironment (TME).
• The elevation of IL10RA (IL-10 receptor alpha) suggests increased responsiveness to IL-10, a potent immunosuppressive cytokine often abundant in the TME. This could contribute to T cell anergy or exhaustion, dampening effective anti-tumor immunity in PDAC https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3653112/.
• CD28, a critical co-stimulatory molecule, is also upregulated. This could signify ongoing T cell activation in response to tumor antigens, but within the suppressive PDAC TME, this activation might be insufficient or lead to exhaustion rather than effective anti-tumor responses.
• Upregulation of IL6R indicates that CD4+ T cells are responsive to IL-6, a pro-inflammatory cytokine frequently elevated in PDAC, which can promote tumor progression and immunosuppression, including driving Th17 differentiation https://pubmed.ncbi.nlm.nih.gov/22420803/.
• Adhesion molecules like ICAM2 and ITGB7, along with SPN (CD43), suggest altered T cell trafficking, extravasation into the tumor, or retention within the TME.
• The expression of GPR65 (TDAG8), a pH-sensing receptor, might indicate T cell adaptation to the acidic conditions often found within solid tumors, influencing their function and survival in this harsh environment https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3186105/.
• Significance of Surfaceome Markers: The identification of these specific surface-expressed proteins is particularly valuable, as they represent accessible molecular targets for manipulating T cell function or for diagnostic applications.

Clinical or Translational Implications

The distinct surfaceome marker profiles observed in CD4+ T cells from PDAC versus normal pancreas tissue offer significant potential for clinical translation:

• Biomarker Development: The identified PDAC-specific markers, such as IL10RA, IL6R, CD28, GPR65, ICAM2, and ITGB7, are excellent candidates for diagnostic or prognostic biomarkers. Their expression on tumor-infiltrating CD4+ T cells could help differentiate diseased tissue from healthy tissue, or their levels might correlate with disease severity, progression, or therapeutic response.
• Novel Therapeutic Targets: Given their surface localization, these markers are attractive candidates for targeted immunotherapeutic strategies in PDAC.
• Blocking immunosuppressive pathways: Inhibitors or antibodies targeting IL10RA or IL6R could be developed to counteract the immunosuppressive effects of IL-10 and IL-6 in the PDAC TME, potentially reactivating anti-tumor immunity.
• Modulating T cell function: Strategies to modulate CD28 activity or target GPR65 (e.g., to exploit its pH-sensing properties) could offer novel approaches to enhance the anti-tumor function of CD4+ T cells in PDAC.
• Targeting cell trafficking: Altering the function of adhesion molecules like ICAM2 or ITGB7 could influence T cell infiltration or retention within the tumor.
• Patient Stratification: The observed heterogeneity in marker expression across PDAC samples suggests that these markers could be utilized to stratify patients based on the specific immune phenotype of their T cells. This stratification could guide personalized treatment plans, directing patients to therapies most likely to be effective against their particular tumor immune profile.
• Experimental Validation: These findings provide a strong foundation for further experimental validation. Techniques like multiparametric flow cytometry, mass cytometry, or immunohistochemistry on patient tissue samples could confirm protein expression patterns. Functional studies in *in vitro* models or *in vivo* animal models would be crucial to elucidate the precise roles of these markers in PD4+ T cell biology and their impact on anti-tumor immune responses in PDAC.

17. Dysregulation of Cell Cycle Pathway Genes in Pancreatic Ductal Adenocarcinoma (PDAC) Ductal Cells

[Analysis Visualization Results]...

Analysis Overview

This analysis investigates the expression patterns of a predefined set of cell cycle pathway genes within Ductal cells from Pancreatic Ductal Adenocarcinoma (PDAC) tissue compared to Ductal cells from adjacent normal (Adj_normal) pancreatic tissue. Ductal cells are identified as the tumor origin cell type in PDAC. The primary goal is to identify genes with statistically significant differences in expression fraction between these two conditions, providing insights into the proliferative state and regulatory dysfunctions in PDAC.

Visual Summary

The visualization displays 24 box plots, each representing the "Expressing cell fraction (sample)" for a specific cell cycle pathway gene in Ductal cells. The expression fractions are compared between 'Adj_normal' (blue boxes) and 'PDAC' (orange boxes) conditions.

Key observations from the plots include:

• Widespread Upregulation in PDAC: For all 24 displayed genes, the expressing cell fraction is significantly higher in PDAC samples compared to Adj_normal samples. This is indicated by the higher median and overall distribution of data points for PDAC (orange boxes) and the accompanying p-values, which are all well below the 0.05 significance cutoff (e.g., ANAPC1: p=6.61e-05, MCM7: p=9.42e-05, TGFB1: p=0.000386, MYC: p=0.00852).
• Increased Proliferation Signature: Genes such as MCM7 (Mini-chromosome maintenance complex component 7), ORC2 (Origin Recognition Complex Subunit 2), CCND1 (Cyclin D1), CDK7 (Cyclin Dependent Kinase 7), ANAPC1 (Anaphase Promoting Complex Subunit 1), CDC16, CDC23, and FZR1 (Cell Division Cycle 20 Homolog) show clear and substantial increases in the fraction of expressing cells in PDAC. These genes are directly involved in DNA replication and mitotic progression.
• Dysregulation of Regulatory Elements: Genes involved in cell cycle regulation and checkpoint control like WEE1 (WEE1 G2 Checkpoint Kinase), SFN (Stratifin, also known as 14-3-3 sigma), CDC25B (Cell Division Cycle 25B), ATR (ATM and Rad3 Related), TP53 (Tumor Protein P53), RB1 (Retinoblastoma Transcriptional Repressor 1), and CDKN2D (Cyclin Dependent Kinase Inhibitor 2D) also show increased expression fractions.
• Signaling Pathway Components: Components of crucial signaling pathways like TGFB1 (Transforming Growth Factor Beta 1), SMAD2, SMAD3, GSK3B (Glycogen Synthase Kinase 3 Beta), MYC (MYC Proto-Oncogene, BHLH Transcription Factor), CREBBP (CREB Binding Protein), and ZBTB17 (Zinc Finger And BTB Domain Containing 17) are also notably upregulated.

The box plots, overlaid with individual sample data points (stripplot), clearly illustrate the inter-sample variability while robustly demonstrating the statistically significant shift towards higher expression fractions in the PDAC condition for these cell cycle-related genes.

Biological Interpretation

The observed widespread upregulation of cell cycle pathway genes in Ductal cells from PDAC, compared to adjacent normal Ductal cells, strongly indicates an accelerated and dysregulated cell cycle, a hallmark of cancer. Given that Ductal cells are the tumor origin cell type in PDAC, these changes are direct indicators of malignant transformation and uncontrolled proliferation.

• Accelerated Cell Cycle Progression: The increased expression fraction of genes like MCM7 and ORC2, which are integral to DNA replication initiation https://www.genecards.org/cgi-bin/carddisp.pl?gene=MCM7, suggests enhanced DNA synthesis activity. Similarly, upregulation of cyclins (e.g., CCND1) and cyclin-dependent kinases (e.g., CDK7) drives progression through different phases of the cell cycle, promoting cell division https://www.genecards.org/cgi-bin/carddisp.pl?gene=CCND1. Genes like ANAPC1, ANAPC10, CDC16, CDC23, and FZR1 are components or regulators of the Anaphase-Promoting Complex/Cyclosome (APC/C), which is essential for mitotic progression and chromosome segregation https://pubmed.ncbi.nlm.nih.gov/12474136/. Their upregulation suggests active and potentially uncontrolled mitotic division.
• Dysregulation of Checkpoints and DNA Damage Response: The simultaneous upregulation of genes involved in cell cycle checkpoints and DNA damage response, such as ATR, TP53, SFN, WEE1, RB1, and CDKN2D, requires nuanced interpretation. While TP53 and RB1 are well-known tumor suppressors https://www.genecards.org/cgi-bin/carddisp.pl?gene=TP53, their increased expression could reflect several scenarios in cancer:

1. Response to Oncogenic Stress/DNA Damage: Actively proliferating cancer cells often experience replication stress and accumulate DNA damage. Upregulation of DNA damage sensors (e.g., ATR) or effectors (e.g., TP53) could be a cellular response attempting to halt proliferation or induce apoptosis, a response that is typically bypassed in cancer cells due to mutations or inactivation of these pathways.
2. Mutated/Inactive Proteins: High expression of a non-functional or mutated tumor suppressor (e.g., mutant TP53) can paradoxically contribute to oncogenesis, as the protein may lose its tumor-suppressive functions while still being transcribed.
3. Compensatory Mechanisms: In highly proliferative contexts, cells might attempt to upregulate inhibitors (e.g., CDKN2D, WEE1) as a compensatory mechanism, even if the overall oncogenic drive overrides their effects.

• Oncogenic Signaling Activation: The increased expression of MYC, a potent oncogene that drives cell proliferation, growth, and metabolism, is a classic hallmark of cancer https://www.genecards.org/cgi-bin/carddisp.pl?gene=MYC. The upregulation of TGFB1, SMAD2, and SMAD3 suggests active TGF-beta signaling. While TGF-beta acts as a tumor suppressor in early stages, it often switches roles to promote tumor progression, invasion, and immune evasion in advanced PDAC https://pubmed.ncbi.nlm.nih.gov/30670860/. GSK3B and CREBBP are also involved in various signaling pathways that impact cell cycle control and proliferation, and their dysregulation can contribute to tumorigenesis.

Overall, the data robustly demonstrates that a significantly higher proportion of Ductal cells in PDAC samples are engaged in active cell cycle processes, driven by the coordinated upregulation of a broad range of cell cycle drivers, regulators, and related signaling components.

Clinical or Translational Implications

The findings have significant clinical and translational implications for PDAC:

• Therapeutic Targets: The consistently upregulated cell cycle genes represent promising therapeutic targets for PDAC. Inhibitors against CDKs (e.g., CDK4/6 inhibitors), ATR inhibitors, or mitotic regulators are already in various stages of clinical development or approved for other cancers. Targeting these pathways could help curb the uncontrolled proliferation characteristic of PDAC.
• Biomarker Development: The increased expression fraction of these cell cycle genes in Ductal cells could serve as diagnostic or prognostic biomarkers for PDAC. Detecting elevated levels of these genes in patient samples, possibly via liquid biopsy or tissue biopsies, could aid in early diagnosis, monitoring disease progression, or predicting response to therapy.
• Understanding Disease Mechanisms: This analysis provides fundamental insights into the molecular mechanisms driving PDAC pathogenesis, emphasizing the profound dysregulation of cell cycle control in the tumor-initiating Ductal cells. This understanding is crucial for developing more effective and targeted treatment strategies.

18. Ductal Cell Gene Ontology Analysis: Insights into Pancreatic Homeostasis and Pancreatic Ductal Adenocarcinoma (PDAC) Pathobiology

[Analysis Visualization Results]...

Analysis Overview

This analysis investigates the biological pathways and processes (Gene Ontology, GSA) that are significantly upregulated in Ductal cells under different conditions: adjacent normal tissue (Adj_normal), PDAC tumor tissue, and based on their ploidy status (Diploid). Ductal cells are the presumed cell of origin for PDAC. The analysis, presented as bar plots, highlights key functional shifts associated with disease and cellular state in this critical cell type. The GSA_up designation indicates that the terms shown are enriched for genes upregulated in the target group compared to the comparison group.

Visual Summary

The visualizations provide a clear comparison of enriched pathways in Ductal cells across three distinct contexts:

1. Ductal cell: Adj_normal_vs_others: This plot displays pathways upregulated in Ductal cells from adjacent normal pancreatic tissue compared to Ductal cells from other conditions (primarily PDAC). The top enriched terms are heavily skewed towards pancreatic digestive functions and diverse metabolic processes, such as "Pancreatic secretion", "Protein digestion and absorption", and various amino acid and lipid metabolism pathways. The p-values are highly significant, indicating robust enrichment of these physiological functions.
2. Ductal cell: Diploid_vs_others: This plot shows pathways upregulated in Ductal cells classified as Diploid (vs. Aneuploid). It features a shorter list of significantly enriched terms, primarily related to immune signaling and fundamental growth/survival pathways. Notable terms include "Complement and coagulation cascades", "Chemokine signaling pathway", "Leukocyte transendothelial migration", and "PI3K-Akt signaling pathway". The -log(p-val) values are substantial, although the list is less extensive than for the other two conditions.
3. Ductal cell: PDAC_vs_others: This plot presents a substantially longer list of pathways upregulated in Ductal cells from PDAC tissue compared to adjacent normal tissue. The enriched terms encompass a wide array of cellular stress responses, protein processing and degradation, fundamental cellular processes like "Autophagy" and "Spliceosome", metabolic alterations, and numerous pathways directly or indirectly linked to various cancers and infectious diseases. Prominent cancer-related pathways include "mTOR signaling pathway", "Cellular senescence", "ErbB signaling pathway", "TNF signaling pathway", "Pathways in cancer", and directly "Pancreatic cancer".

Biological Interpretation

Adjacent Normal Ductal Cells: Maintaining Pancreatic Homeostasis

Ductal cells from adjacent normal pancreatic tissue exhibit a robust upregulation of pathways directly associated with their physiological roles. The top terms, such as "Pancreatic secretion" and "Protein digestion and absorption", underscore their primary function in digestive enzyme transport and fluid/bicarbonate secretion, essential for normal digestion. The enrichment of various amino acid, lipid, and carbohydrate metabolic pathways (e.g., "Arginine and proline metabolism", "Fatty acid degradation") indicates an active and diverse metabolic profile crucial for maintaining cellular energy and building blocks. This suggests that even in proximity to a tumor, these adjacent normal Ductal cells largely retain their specialized physiological functions, highlighting a stark contrast with the malignant phenotype.

Diploid Ductal Cells: Immune Engagement and Early Signaling

The finding that Diploid Ductal cells show upregulation of "Complement and coagulation cascades", "Chemokine signaling pathway", and "Leukocyte transendothelial migration" suggests an active role in immune responses. Diploid cells, potentially representing a less transformed or earlier stage of malignancy compared to aneuploid cells, might be more engaged in immune surveillance or inflammatory processes. The significant enrichment of the "PI3K-Akt signaling pathway" in diploid cells is also noteworthy. This pathway is a central regulator of cell growth, proliferation, survival, and metabolism, and its activation can be an early event in cellular transformation or a mechanism employed by a subset of non-aneuploid tumor cells or reactive stromal cells to promote survival. PubMed search: PI3K-Akt signaling cancer early stages

PDAC Ductal Cells: A Landscape of Malignant Reprogramming and Stress

In contrast, Ductal cells from PDAC tumors display a dramatic shift in their functional landscape. The extensive list of upregulated pathways points to a highly reprogrammed and stressed cellular state characteristic of malignancy:

• Protein Homeostasis and Stress Response: Terms like "Protein processing in endoplasmic reticulum", "Ubiquitin mediated proteolysis", "Autophagy", "Spliceosome", and "Lysosome" indicate a heightened demand for protein synthesis, folding, degradation, and organelle quality control. This is a common feature in rapidly proliferating cancer cells that experience metabolic stress and increased protein turnover. PubMed search: ER stress cancer protein homeostasis
• Altered Metabolism and Growth Signaling: Pathways such as "mTOR signaling pathway", "AMPK signaling pathway", "Insulin signaling pathway", and "Lipid and atherosclerosis" highlight significant metabolic reprogramming, a hallmark of cancer that supports rapid growth and adaptation to nutrient-deprived microenvironments. PubMed search: mTOR signaling pancreatic cancer metabolism
• Cellular Senescence and Survival: The upregulation of "Cellular senescence" suggests a mechanism by which some tumor cells might attempt to halt proliferation or undergo stress-induced changes, potentially contributing to the tumor microenvironment.
• Immune Evasion and Inflammation: While many infection-related terms (e.g., "Salmonella infection", "Human papillomavirus infection") might reflect broad inflammatory or stress responses rather than actual infection, they underscore the complex immune interactions and inflammatory milieu within the tumor. The "TNF signaling pathway" also points to chronic inflammation.
• Direct Cancer Pathways: The direct enrichment of "Pancreatic cancer" and "Pathways in cancer" terms provides strong confirmation of the malignant phenotype. Other cancer-associated pathways like "ErbB signaling pathway" (often involving EGFR, a target in some cancers) and "Cell cycle" further support the aggressive proliferative nature of these cells.

Clinical or Translational Implications

These GSA results provide crucial insights into the molecular differences between healthy and cancerous Ductal cells, which can have significant clinical implications:

1. Biomarker Discovery: The distinct sets of upregulated pathways in adjacent normal vs. PDAC Ductal cells could serve as sources for novel diagnostic or prognostic biomarkers. For instance, genes highly specific to pancreatic secretion might be downregulated in early PDAC, while genes in the ER stress or mTOR pathways could be upregulated.
2. Therapeutic Targets: The activated pathways in PDAC Ductal cells, such as mTOR, ErbB, and those involved in protein processing (e.g., proteasome, ER stress), represent potential therapeutic targets. Inhibiting these pathways could disrupt the survival and proliferation of malignant Ductal cells. PubMed search: therapeutic targets PDAC mTOR ErbB
3. Understanding Tumor Heterogeneity: The distinction between Diploid and Aneuploid Ductal cells suggests different biological underpinnings based on ploidy status. If Diploid Ductal cells represent an earlier stage or specific subpopulation, understanding their unique pathway activation (e.g., PI3K-Akt signaling, immune-related pathways) could lead to targeted interventions for specific tumor cell subsets or early disease.
4. Microenvironment Interactions: The involvement of immune-related pathways in Diploid Ductal cells and general inflammatory responses in PDAC cells highlights the critical interplay between tumor cells and the immune microenvironment. Modulating these interactions could be a strategy for immunotherapy or combination therapies in PDAC.

19. 췌장암(PDAC) 미세환경 내 세포 유형별 유전자 세트 농축 분석 (GSEA)

[Analysis Visualization Results]...

Analysis Overview

제공된 점도표(dot plot)는 단일 세포 RNA 시퀀싱 데이터를 기반으로 다양한 세포 유형에서 관찰되는 유전자 세트 농축 분석(Gene Set Enrichment Analysis, GSEA) 결과를 시각화한 것입니다. 각 점은 특정 세포 유형 및 조건(x축)에서 특정 경로(y축)의 농축 정도를 나타냅니다. 점의 크기는 통계적 유의성(-log(p-value))을, 색상은 정규화된 농축 점수(Normalized Enrichment Score, NES)를 나타냅니다. 이 분석은 췌장암(PDAC) 조건과 인접 정상(Adj_normal) 또는 다른 세포 유형(others)을 비교하여 PDAC 미세환경 내 각 세포 유형의 기능적 변화를 파악하는 데 중점을 둡니다.

참고: 사용자 요청 시 cmap은 'RdBu_r'로, n_pws_to_show는 120으로 설정되었으나, 실제 적용된 파라미터는 cmap='Reds'와 n_pws_to_show=60입니다. 이에 따라 음의 NES 값은 시각화되지 않았을 가능성이 있으며, 표시되는 경로의 수가 제한되었습니다.

Visual Summary

점도표는 PDAC 미세환경 내 다양한 세포 유형에서 활성화되거나 하향 조절되는 주요 생물학적 경로들을 명확하게 보여줍니다.

• 점 크기(유의성): 점이 클수록 해당 경로의 농축이 통계적으로 매우 유의함을 나타냅니다 (낮은 p-value). 많은 경로가 다양한 세포 유형에서 큰 점으로 나타나, PDAC 상태에서 광범위한 유전자 발현 변화가 있음을 시사합니다.
• 점 색상(NES): 색상 막대에서 붉은색이 NES 0.0부터 2.5까지를 나타내며, 어두운 붉은색일수록 더 높은 양의 NES 값을 의미합니다. 이는 해당 경로가 활성화(upregulation)되었음을 뜻합니다. 전반적으로 많은 경로에서 양의 NES 값이 관찰되어, PDAC 환경에서 다양한 생물학적 과정이 상향 조절됨을 보여줍니다.
• 경로 분포: 특정 경로는 여러 세포 유형에서 일관되게 농축되는 반면, 일부 경로는 특정 세포 유형이나 조건에서만 독특하게 농축됩니다. 예를 들어, "Protein processing in endoplasmic reticulum", "Ubiquitin mediated proteolysis", "Lysosome" 등은 다양한 세포 유형에서 공통적으로 높은 NES와 유의성을 보입니다.
• 세포 유형별 특징:
• Ductal cell (PDAC_vs_others): 종양 기원 세포인 Ductal cell에서 "ErbB signaling pathway", "Wnt signaling pathway", "Insulin signaling pathway" 등 암 관련 핵심 신호 경로와 "Protein processing in endoplasmic reticulum", "Ubiquitin mediated proteolysis" 등 단백질 항상성 및 스트레스 관련 경로가 강하게 농축되어 있습니다.
• Macrophage (PDAC_vs_others): 대식세포에서는 "Antigen processing and presentation", "Phagosome", "Fc gamma R-mediated phagocytosis", "IL-17 signaling pathway", "TNF signaling pathway" 등 면역 및 염증 관련 경로들이 두드러지게 농축됩니다.
• T cell CD4+ 및 CD8+ (PDAC_vs_others): T 세포들에서는 "T cell receptor signaling pathway", "IL-17 signaling pathway", "TNF signaling pathway" 등 T 세포 활성화 및 염증 반응 관련 경로가 농축되어 있습니다.
• Endothelial cell (PDAC_vs_others): 혈관내피세포에서는 "Fluid shear stress and atherosclerosis" 및 일부 대사 관련 경로가 눈에 띄게 농축되어 혈관 재형성 및 대사 변화를 시사합니다.

Biological Interpretation

PDAC는 복잡한 종양 미세환경(Tumor Microenvironment, TME)을 특징으로 하며, 본 GSEA 결과는 각 세포 유형이 PDAC 병리에서 수행하는 역할을 조명합니다.

1. Ductal Cell (종양 세포)의 기능적 재프로그래밍:

• PDAC Ductal cell에서 "ErbB signaling pathway", "Wnt signaling pathway", "Insulin signaling pathway"의 활성화는 췌장암 발병 및 진행의 주요 동인임을 시사합니다. 이들 경로는 세포 성장, 증식, 생존 및 전이에 중요하게 관여합니다. 참고: Wnt Signaling Pathway - GeneCards
• "Protein processing in endoplasmic reticulum", "Ubiquitin mediated proteolysis", "Lysosome", "RNA degradation"의 농축은 종양 세포의 높은 대사 활동과 함께 ER 스트레스, 단백질 품질 관리 시스템의 활성화를 반영합니다. 이는 암세포의 빠른 증식과 관련된 단백질 합성 증가 및 스트레스 적응 메커니즘을 나타낼 수 있습니다.

2. 면역 세포의 활성화 및 재편:

• Macrophage (대식세포): PDAC 대식세포에서 "Antigen processing and presentation", "Fc gamma R-mediated phagocytosis", "Phagosome", "Lysosome" 등의 경로 활성화는 이들이 적극적으로 주변 환경과 상호작용하며 세포 잔해 처리 및 항원 제시 기능을 수행하고 있음을 보여줍니다. "IL-17 signaling pathway"와 "TNF signaling pathway"의 농축은 종양 관련 대식세포(TAMs)가 종종 보이는 염증 유발 및 면역억제 환경 조성에 기여하는 복합적인 역할을 시사합니다. 참고: Tumor-associated macrophages - PubMed Search
• T cell CD4+ 및 CD8+: T 세포 수용체 신호 경로의 활성화는 T 세포가 항원에 반응하고 있음을 나타내지만, 이것이 효과적인 항종양 반응인지 또는 기능이 저하된(exhausted) 상태인지는 추가적인 유전자 발현 분석이 필요합니다. "IL-17 signaling pathway" 및 "TNF signaling pathway"의 농축은 TME 내 염증 반응이 중요함을 강조합니다.
• NK cell (자연살해 세포): "Natural killer cell mediated cytotoxicity" 경로의 활성화는 NK 세포가 종양 세포에 대한 직접적인 세포독성 기능을 시도하고 있음을 보여줍니다. 이는 PDAC의 면역 회피 전략에도 불구하고 일부 NK 세포가 활성화되어 있음을 시사합니다.

3. 미세환경 내 스트로마 및 혈관 세포의 반응:

• Endothelial cell (혈관내피세포): "Fluid shear stress and atherosclerosis" 경로의 농축은 PDAC TME 내에서 발생하는 혈관 신생 및 혈관 재형성과 관련된 형태학적 및 기능적 변화를 반영합니다. 종양 혈관은 종종 비정상적인 구조와 기능을 가지며, 이는 산소 및 영양분 공급, 약물 전달에 영향을 미칩니다.
• Smooth muscle cell (평활근 세포): 이 세포들에서 발견되는 ER 스트레스 및 대사 관련 경로는 스트로마 내 평활근 세포가 종양 성장에 반응하여 활성화되고, 세포 외 기질 재형성 및 종양 침윤에 기여할 수 있음을 나타냅니다.

4. 광범위한 세포 스트레스 및 대사 재편:

• "Protein processing in endoplasmic reticulum", "Lysosome", "Peroxisome"과 같은 경로는 Ductal, Acinar, Endothelial, Macrophage, Smooth muscle cell 등 여러 세포 유형에서 공통적으로 농축되어 있습니다. 이는 PDAC TME 전체에 걸쳐 세포 스트레스 반응, 단백질 및 지질 대사의 변화가 광범위하게 일어나고 있음을 시사합니다. 이러한 대사 재편은 암세포의 증식 요구를 충족시키고 주변 세포의 기능을 변화시키는 데 중요합니다.

Clinical or Translational Implications

이 GSEA 결과는 PDAC의 치료 전략 개발에 중요한 통찰력을 제공할 수 있습니다.

1. 새로운 치료 표적 발굴: Ductal cell에서 활성화된 "ErbB", "Wnt", "Insulin" 신호 경로들은 이미 암 치료의 표적으로 연구되어 왔으며, PDAC 특이적 표적 치료제 개발의 근거가 될 수 있습니다. 특히, 특정 경로 활성화가 PDAC 진행에 결정적인 역할을 하는 경우, 이를 억제하는 약물은 효과적인 치료 전략이 될 수 있습니다.
2. 면역 치료 전략 최적화: Macrophage, T cell, NK cell에서 나타나는 면역 및 염증 관련 경로의 활성화는 PDAC의 면역 환경이 복잡함을 보여줍니다. TAMs의 염증 유발 및 면역억제 기능 조절, T 세포의 항종양 반응 강화, NK 세포의 세포독성 활성 유도는 PDAC에 대한 면역 치료 효능을 높이는 데 기여할 수 있습니다. 참고: Cancer immunotherapy - PubMed Search
3. TME 조절을 통한 치료: Endothelial cell의 혈관 재형성 관련 경로 활성화 및 스트로마 세포의 대사 변화는 혈관 신생 억제제 또는 스트로마 조절 약물과 같은 TME 표적 치료 전략을 모색할 수 있게 합니다. TME를 정상화하는 접근 방식은 약물 전달을 개선하고 종양 성장을 억제할 수 있습니다.
4. 바이오마커 개발: 특정 세포 유형에서 높은 유의성과 NES로 농축되는 경로들은 PDAC의 진단, 예후 예측 또는 치료 반응 예측을 위한 바이오마커로 활용될 가능성이 있습니다. 예를 들어, Ductal cell의 ER 스트레스 관련 유전자 발현 패턴은 특정 치료에 대한 반응성을 예측할 수 있습니다.

20. Discussion

The single-cell RNA sequencing analysis of pancreatic tissue provides a granular view of the profound cellular and molecular reprogramming that characterizes Pancreatic Ductal Adenocarcinoma (PDAC) compared to adjacent normal tissue. A central finding is the clear identification of malignant Ductal cells, the presumed tumor origin cell type, which exhibit extensive genomic instability and a highly proliferative state.

Malignant Ductal Cell Transformation: UMAP visualizations clearly separate PDAC cells from adjacent normal counterparts, primarily driven by the emergence of a large aneuploid Ductal cell population in PDAC samples. This aneuploidy is corroborated by specific Ductal cell ploidy analysis, showing a dramatic shift from diploidy to aneuploidy in the majority of PDAC samples. Copy Number Variation (CNV) analysis further elucidates the genomic chaos, revealing recurrent amplifications on chromosomes 1q, 7p (including the *EGFR* locus), 8q (implicating *EIF3E, GSDMD*), 17q, 19q, and 20q in PDAC Ductal cells. Functionally, these malignant Ductal cells show a widespread upregulation of cell cycle pathway genes (e.g., *MCM7, ORC2, CCND1, MYC, TP53*), indicating uncontrolled proliferation and dysregulation of cell cycle checkpoints. Gene Ontology (GSA) and Gene Set Enrichment Analysis (GSEA) further confirm a dramatic metabolic reprogramming towards high protein turnover, ER stress, autophagy, and activation of oncogenic pathways such as ErbB, Wnt, mTOR, and insulin signaling. The surfaceome of these malignant Ductal cells is also distinct, featuring high expression of pro-tumorigenic and invasion-promoting markers like *ERBB2, ERBB3, MET, SLC2A1, CDCP1, PLAUR, CEACAM1, TSPAN1*.

Remodeling of the Tumor Microenvironment (TME): The cellular composition of the PDAC TME undergoes significant restructuring. Population analysis reveals a drastic reduction of normal Acinar cells and a concomitant increase in Ductal, Fibroblast, Stellate, and Macrophage populations in PDAC. The stromal compartment is notably expanded, with Fibroblasts and Stellate cells showing activated myofibroblast-like phenotypes, characterized by markers like *DCN, LUM, COL1A1, PDGFRA, FAP*, and *ACTA2*, reflecting the pronounced desmoplastic reaction.

Immune Landscape Shifts towards Immunosuppression: The immune cell compartment in PDAC is profoundly altered. While innate lymphoid cells (ILCs) and NK cells show increased proportions, the adaptive T cell landscape is skewed towards immunosuppression. Specifically, there's a significant decrease in cytotoxic T cells (*T_Cyto*) and a notable increase in regulatory T cells (*Treg*), naive T cells (*T_Naive*), and various helper T cell subsets (Th9, Th17, Tfh, Th2, Th22) in PDAC. CD4+ T cells within PDAC upregulate markers such as *IL10RA, CD28, IL6R, GPR65*, suggesting adaptation to the suppressive TME. Macrophages are consistently enriched and reprogrammed in PDAC, expressing pro-tumorigenic surface markers like *SIRPA, ITGAX, BSG, IL6R, TGFBR2, LILRB2*, and showing activation of antigen processing, phagosome, IL-17, and TNF signaling pathways via GSEA. These findings underscore their role in fostering an immunosuppressive environment.

Dysregulated Intercellular Communication: Cell-cell interaction (CCI) analyses unveil a complex and dysregulated communication network in PDAC compared to adjacent normal tissue. In PDAC, 'Diploid Ductal' cells actively engage in extensive crosstalk with Macrophages, and exhibit self-interactions. Key immunosuppressive interactions include TGFB1-TGFBR1 (Macrophage-Ductal, promoting fibrosis and T cell anergy), LGALS9-HAVCR2 (Galectin-9-TIM-3) (Ductal-Macrophage, inducing T cell exhaustion), and FEBP1-LILRB4 (Ductal-Macrophage, promoting immunosuppressive TAMs). Furthermore, macrophage-T cell CD8+ interactions are highly activated, involving immune evasion axes like SIRPA-CD47 (Macrophage-T cell, 'don't eat me' signal) and HLA-E-CD94:NKG2A (Macrophage-T cell, suppressing cytotoxicity), as well as FASLG-FAS (T cell-Macrophage, inducing T cell apoptosis). Pro-tumorigenic stromal remodeling is supported by SPP1-integrin (Macrophage-Ductal, promoting invasion) and PDGFB-PDGFRB (Ductal-Macrophage, activating fibroblasts). These altered communication patterns represent a central mechanism by which the PDAC TME promotes tumor growth and evades immune surveillance.

Hypotheses:

1. Malignant Ductal cells in PDAC actively drive immunosuppression and stromal remodeling through specific ligand-receptor interactions, such as TGFB1-TGFBR1 and PDGFB-PDGFRB, thereby creating a permissive microenvironment for tumor progression.
2. The observed shift in T cell subsets (decreased cytotoxic T, increased Treg/naive/Th subsets) is a direct consequence of immune checkpoints and immunosuppressive signals originating from malignant Ductal cells and tumor-associated macrophages, leading to T cell exhaustion and ineffective anti-tumor responses.
3. The widespread upregulation of cell cycle genes and metabolic reprogramming pathways in PDAC Ductal cells is sustained by specific surface receptors (e.g., ERBB2, MET, SLC2A1) and represents a key vulnerability for therapeutic targeting of tumor cell proliferation and survival.
4. Inter-patient heterogeneity in Ductal cell CNV profiles and immune cell composition within PDAC samples dictates differential responses to standard and emerging therapies, highlighting the need for personalized approaches.

Potential therapeutic targets:

1. EGFR (Epidermal Growth Factor Receptor): EGFR signaling promotes cell proliferation, survival, and metastasis. Its amplification is a known oncogenic driver. Evidence: Recurrent amplification of the 7p13:7q21.11 locus (containing EGFR) was observed in Ductal cells from 45% of PDAC samples (CNV analysis, Image 5). GSA and GSEA also highlighted the ErbB signaling pathway as upregulated in PDAC Ductal cells (Image 23, Image 24). Validation: Test EGFR inhibitors (e.g., Erlotinib) in PDAC cell lines or organoids with confirmed EGFR amplification. Validate *in vivo* efficacy in patient-derived xenograft (PDX) models. Screen patient cohorts for EGFR amplification and correlate with response to EGFR-targeted therapies.
2. TGF-beta signaling (via TGFBR1): TGF-beta is a potent immunosuppressive and pro-fibrotic cytokine, driving immune evasion, T cell anergy, and desmoplastic stroma formation in PDAC. Evidence: Highly significant and expressed TGFB1-TGFBR1 interactions were observed between Macrophages and Ductal cells in PDAC, but not prominently in Adj_normal tissue (CCI analysis, Image 12). GSA also showed upregulation of TGFB1 in PDAC Ductal cells (Image 20) and the mTOR signaling pathway, which can be influenced by TGF-beta (Image 23). Validation: Utilize TGF-beta inhibitors (e.g., fresolimumab, galunisertib) in PDAC animal models to assess impact on tumor growth, immune cell infiltration, T cell function, and stromal density. Evaluate combination with chemotherapy or immunotherapy. Monitor TGFB1 expression and TGF-beta pathway activation in patient biopsies.
3. SIRPA-CD47 axis: CD47 acts as a 'don't eat me' signal, enabling tumor cells to evade macrophage phagocytosis. Targeting SIRPA on macrophages or CD47 on tumor cells can enhance anti-tumor immunity. Evidence: The SIRPA-CD47 interaction between Macrophages and T CD8+ cells (and likely other cell types including tumor cells, though not explicitly shown from tumor cells) was significantly and robustly activated in PDAC samples compared to Adj_normal (Condition-specific CCI, Image 15). Macrophage-specific surfaceome marker analysis showed upregulation of SIRPA in PDAC macrophages (Image 17). Validation: Test anti-CD47 or anti-SIRPA antibodies in PDAC patient-derived organoids or xenografts to assess their impact on macrophage-mediated phagocytosis and tumor clearance. Combine with checkpoint inhibitors to evaluate synergistic effects. Analyze SIRPA and CD47 expression in patient tumors by immunohistochemistry and correlate with clinical outcomes.
4. ERBB2/HER2 and MET (Receptor Tyrosine Kinases): ERBB2 and MET are receptor tyrosine kinases frequently dysregulated in cancers, promoting aggressive tumor growth, survival, and invasion. Their surface localization makes them excellent targets. Evidence: ERBB2 and MET were among the most highly upregulated surfaceome markers on malignant Ductal cells in PDAC, with high mean expression and prevalence across tumor samples (Ductal cell condition-specific markers, Image 16). GSEA confirmed activation of the ErbB signaling pathway in PDAC Ductal cells (Image 24). Validation: Develop or evaluate existing antibodies or small molecule inhibitors targeting ERBB2 (e.g., Trastuzumab) or MET in PDAC cell lines and organoids expressing these markers. Investigate efficacy of antibody-drug conjugates (ADCs) or bispecific antibodies. Assess potential for patient stratification based on ERBB2/MET expression levels.
5. IL-6 Receptor (IL6R): IL-6 is a pro-inflammatory cytokine promoting tumor growth, metastasis, and immunosuppression in PDAC. Targeting its receptor on immune cells like macrophages and T cells can disrupt this signaling. Evidence: IL6R was significantly upregulated as a surfaceome marker on Macrophages (Image 17) and CD4+ T cells (Image 19) in PDAC samples compared to adjacent normal tissue, suggesting these cells are highly responsive to IL-6 signaling within the TME. GSEA also highlighted IL-17 signaling pathway and TNF signaling pathway in macrophages and T cells, which often cross-talk with IL-6 (Image 24). Validation: Evaluate anti-IL6R antibodies (e.g., Tocilizumab) in PDAC models. Assess effects on macrophage polarization, T cell function, and overall tumor growth. Study combination with other immunotherapies. Correlate IL6R expression with patient response to IL-6 blockade in clinical trials.

Follow-up validation ideas:

1. Perform spatial transcriptomics or multiplexed immunostaining on PDAC patient tissues to map the precise localization and interaction of identified cell types (e.g., aneuploid Ductal cells, specific macrophage subsets, T cell subsets) and key ligand-receptor pairs (e.g., SIRPA-CD47, LGALS9-HAVCR2, TGFB1-TGFBR1) within the tumor microenvironment.
2. Conduct in vitro co-culture experiments using patient-derived organoids or cell lines of Ductal cells, macrophages, and T cells to functionally validate the identified cell-cell interactions and their impact on immune cell function, tumor cell proliferation, and invasion. Use blocking antibodies or genetic knockdowns for specific ligand-receptor pairs.
3. Utilize flow cytometry or mass cytometry on dissociated PDAC and adjacent normal tissues to quantify the absolute numbers and phenotypic markers (including surfaceome markers like SIRPA, LILRB2, IL10RA, ERBB2, MET, CDCP1) of the identified immune and malignant cell populations, correlating with disease progression or treatment response in larger patient cohorts.
4. Employ CRISPR/Cas9 or RNA interference in PDAC cell lines to perturb key cell cycle genes (e.g., MYC, CCND1) or oncogenic receptors (e.g., ERBB2, MET) and assess their impact on cell proliferation, survival, and sensitivity to anti-cancer drugs in vitro and in vivo models.
5. Validate the identified CNV regions (e.g., *EGFR*, 1q, 8q amplifications) using fluorescence in situ hybridization (FISH) or comparative genomic hybridization (CGH) on spatially preserved tissue sections, correlating these genomic alterations with protein expression and clinical outcomes.

Limitations:

This single-cell RNA-seq analysis provides a snapshot of gene expression and inferred genomic alterations, and while robust statistical methods were applied, it is correlative and does not establish causality. The study primarily relies on transcriptional and inferred CNV data, and protein-level validation would be crucial. The patient cohort, while diverse, may not capture the full spectrum of PDAC heterogeneity, particularly rare cell types or specific disease subtypes. Furthermore, cell-cell interaction inference is based on ligand-receptor expression and requires functional validation in experimental models to confirm biological relevance. The 'Diploid PDAC' samples warrant further investigation to fully characterize their cellular and genomic states. The resolution of CNV analysis from single-cell RNA-seq is inherently lower than dedicated genomic profiling techniques.

21. Query List

1. Show and save UMAPs for condition, sample, major cell type, minor cell type, ploidy_dec, and celltype_subset in two columns.
2. Show major cell type scores on UMAP and save the result.
3. Show and save a marker expression dot plot for celltype_subset. SET target_cell = None and var_group_rotation = 45. Leave all other arguments at their default values.
4. Select Ductal cells, show a CNV heatmap grouped by sample together with a summary of regions with significantly amplified copy numbers, and save the results.
5. Show and save UMAPs of CNV patterns colored by major cell type, minor cell type, ploidy results, condition, and sample in two columns.
6. Show and save a population bar plot of minor cell types.
7. Show and save a population bar plot of T cell subsets.
8. Show and save box plots of T cell subset populations with statistically significant differences between conditions, if any. Choose ncols based on the total number of panels.
9. Select Ductal cells, show their ploidy populations as a bar plot, and save the result.
10. Show and save cell-cell interaction patterns involving Ductal cells, fibroblasts, macrophages, T cells, and other relevant cell types. Select at most 80 cell-cell interactions per group.
11. Select only genes associated with immune checkpoint and cell cycle pathways, show cell-cell interactions for these genes, and save the result.
12. Find cell-cell interactions involving major immune and stromal cells that differ significantly between conditions, show them as a dot plot, and save the result. Set max_n_items_per_group to 60.
13. Extract condition-specific markers for Ductal cells, show their expression as a dot plot, and save the result. Use only surfaceome markers, with at most 50 markers per condition.
14. Extract condition-specific markers for Macrophage, show the markers expression as a dot plot, and save the result. Use only surfaceome markers, with at most 30 markers per condition.
15. Extract condition-specific markers for Fibroblast, show the markers expression as a dot plot, and save the result. Use only surfaceome markers, with at most 30 markers per condition.
16. Extract condition-specific markers for T cell CD4+, show the markers expression as a dot plot, and save the result. Use only surfaceome markers, with at most 30 markers per condition.
17. Among cell cycle pathway genes, select those with statistically significant expression differences between conditions in Ductal cells, show box plots, and save the result. Set max_n_items_to_plot to 24 and choose ncols based on the total number of panels for an overall width-to-height ratio of approximately 2:3.
18. Show and save Gene Ontology (GSA) analysis results for Ductal cells as a bar plot.
19. Show Gene Set Enrichment Analysis results as a dot plot, and save it. Set the color map to RdBu_r and n_pws_to_show to 120.