Bulk RNA-seq deconvolution heterogeneity across paired pancreatic cancer human samples

Introduction: There is great promise in using genomic data to inform individual cancer treatment plans. Assessing intratumor genetic heterogeneity, studies have shown it may be possible to target biopsies to tumor subclones driving disease progression or treatment resistance. Here, we explore if the interpretation of tumor gene expression analysis varies across two specimens from the same patient.

Material and methods: We performed bulk RNA-seq using FFPE samples from 16 patients who also had a previous separate bulk RNA-seq performed and deposited in TCGA. We used three different deconvolution methods to compare cell type proportions for these paired data. We normalized study-specific gene expression values per gene by calculating transcripts per million and adjusted for batch effect across study to compare median expression values. We also compared the reliability of gene expression measurements. We selected KRAS, TP53, SMAD4, and CDKN2A, as the most mutated genes in pancreatic cancer, and CTNNB1, JUN, SMAD3, SMAD7, and TCF7, as these tend to be enriched in pancreatic cancer compared with adjacent normal tissue.

Results: We found that average cell type proportion varied the most between studies (i.e., samples for each patient) for NK and macrophages (using adjusted p-value 0.05/21 = 0.002). For the differential expression analysis, we did not observe significant differences in average expression of any of the selected genes. We observed substantial concordance (kappa = 0.75) only for JUN with low to moderate concordance (i.e., Kappa value 0.25–0.5) for the remaining 8 genes across the two studies.

Discussion: Together, the findings suggest that more than one tumor sample may be needed for effective treatment planning. Any potential difference in observed expression values across the paired samples could be related to the different cell type proportions across the samples. The sample size was small, and each study used different sequencing technologies, so any interpretation should be confirmed with additional studies.

1 Introduction

Pancreatic Ductal Adenocarcinoma (PDAC) is a highly aggressive malignancy with a heterogeneous tumor microenvironment (Oketch et al., 2024). As a result, patients are often diagnosed at a late stage contributing to its high mortality rate and poor treatment response (Grünwald et al., 2021; Hwang et al., 2022). In 2020, approximately 495,773 new cases were diagnosed worldwide, ranking it as the 12th most common malignancy (Cai et al., 2021; Hu H. F. et al., 2021; Li et al., 2024; Rawla et al., 2019). In the United States, the American Cancer Society reported approximately 60,430 new cases and 48,220 deaths in 2021, ranking pancreatic cancer as the third leading cause of cancer death (Cai et al., 2021; Hu H. F. et al., 2021). Similarly, in the European Union, it is projected that approximately 111,500 people will die from pancreatic cancer by the end of 2025. These statistics underscore the urgent need for improved prevention, early detection, and treatment strategies to mitigate the escalating impact of pancreatic cancer on global health (Ying et al., 2025).

The promise and utility of using genomic data to inform individual cancer treatment plans has been longstanding. However, minimal attention has focused on using genomic data to inform biopsy protocols or guide biopsy sampling strategies to enhance the diagnostic and prognostic yield of tissue sampling. There are limited studies evaluating this gap in knowledge. By assessing intratumor genetic heterogeneity, studies in other cancers have shown it may be possible to target biopsies to tumor subclones driving disease progression or treatment resistance (Blanco-Heredia et al., 2024). The tumor microenvironment (TME) in PDAC is a complex biological barrier with multiple components, such as desmoplasia, hypoxia, presence of various cell types, and complex signaling pathways, making treatment challenging (Binkowski et al., 2024; 2025; Giannoukakos et al., 2024; Padwal et al., 2024).

PDAC’s TME complexity necessitates innovative therapeutic strategies that offer potential solutions for targeted drug delivery and modulation of the microenvironment (Nair et al., 2024). CNV studies analyze the expression levels of genes with amplifications or deletions in malignant tissues compared to normal pancreatic tissues to understand the CNV landscape and identify correlations with survival outcomes (Giannoukakos et al., 2024). These variations can influence gene expression, leading to the dysregulation of critical cellular processes and contributing to cancer development and progression (Mazur et al., 2010). Studies employing RNA-Seq,or protein detection have identified potential biomarkers for early detection of PDAC (Butera et al., 2024; Karmakar et al., 2020; Liu R. et al., 2024) These studies collectively demonstrate that PDAC has detectable changes based on changing disease conditions and the TME that if known could inform biopsy and treatment approaches.

Here, we explored the need of targeted biopsy sampling in PDAC by using bulk RNA-seq data to analyze variation gene expression across paired tumor samples from the same group of individuals. Our goal is to assess if we observe any changes in the interpretation of key gene expression data as we analyze RNA-seq data across two samples that should accurately represent the same tumor profile. In other words, are we able to create reliable treatment plans for a patient based on RNA-seq collected from one sample or is there to much variation across samples to do so accurately?

2 Materials and methods

2.1 Sample collection

We performed bulk RNA-seq extracted from FFPE samples of 16 patients who also had bulk RNA-seq performed on a different section from the same tumor and deposited in TCGA. The pipeline and workflow details for the TCGA can be found on their website. (https://docs.gdc.cancer.gov/Data/Bioinformatics_Pipelines/Expression_mRNA_Pipeline/). The second sample for these16 patients was processed using the NovaSeq S4 PE100 with Illumina’s TruSeq Total Stranded RNA prep reagents (https://www.illumina.com/products/by-type/sequencing-kits/library-prep-kits/truseq-stranded-total-rna.html) and used a second DNase treatment to minimize any potential DNA contamination. Table 1 shows select patient characteristics for our study (Table 1). The Institutional Review Board of Mayo Clinic gave approval for this work.

Table 1

Table 1. Select patient characteristics for 16 patients with paired samples.

2.2 RNA-seq data processing

For alignment and quantification of the Mayo RNA-Seq data, we used a pipeline based on HiSat2 and subread (CHURP (Baller et al., 2019), HiSat2 (Kim et al., 2019), and featureCounts (Liao et al., 2014)). The TCGA and Mayo data sets were analyzed independently for deconvolution. For each data set, raw counts were normalized to transcripts per million (tpm). Cell type reference signatures were created from two different published single cell RNA-seq studies: sig1 from GSE229413 and sig2 from GSE205049. Annotations from the original publications were used to define cell types and to subset cells to those from tumor samples (not normal tissue). A final, consolidated reference signature matrix consisting of 23 distinct cell-type profiles was used. We filtered each signature set to remove genes with all zeros in all samples and then filtered each data set to remove any genes with mean expression across all samples that waere less than the overall median gene expression value. This was done to ensure that the most robust genes were kept for the signatures. We kept only the genes that were found in both scRNA-seq signatures to have the same set of genes. We then used granulator (Pfister S et al., 2024) to run deconvolution for each bulk RNA-seq data set with each scRNA-seq reference signature. The granulator R package is designed to run multiple deconvolution algorithms, and we report the non-negative algorithm results for the methods defined by the granulator package as dtangle, nnls, and qprogwc. We used these three different deconvolution methods to compare cell type proportion estimates for both runs of RNA-seq separately.

2.3 Justification of deconvolution methodology and sample size

The use of 23 established cell-type signatures with a limited sample size of 16 tumor samples was a necessary methodological choice. It is important to clarify that cell type deconvolution is not a standard regression model. The 23 cell-type signatures (features) represent fixed, biologically validated reference profiles derived from large external single-cell studies. The deconvolution algorithms (nnls, dtangle, qprogwc) apply constrained linear models (e.g., non-negative and sum-to-one constraints) to estimate the mixing proportions of these known, fixed profiles within the bulk data. These models do not learn or fit the 23 signatures within our samples. We acknowledge that the low sample size limits the power to detect small differences in cell fractions and that the results are susceptible to single-sample variation. Therefore, our findings are presented as preliminary and hypothesis-generating, aiming to establish consistency across three independent deconvolution algorithms and two technical replicates (Mayo and TCGA RNA-seq runs) to demonstrate the robustness of the core immune/stromal shifts.

2.4 Batch correction

We evaluated the impact of batch correction and normalization on median expression. We present in the main manuscript the log2 normalized study-specific gene expression values per gene by calculating tpm across all shared genes and then compare median expression values of key pancreatic cancer genes between studies. We use the ComBat-seq function in the R package sva to adjust for batch effect using an empirical Bayes framework. EdgeR to was used to normalize the batch corrected counts within gene and transformed the values by the log2 function. We selected KRAS, TP53, SMAD4, and CDKN2A, as these are the genes which are commonly mutated in PDAC (Hu J. X. et al., 2021). We also selected CTNNB1, JUN, SMAD3, SMAD7, and TCF7, as these transcription factor enrichment identified genes have been observed to have altered gene expression in PDAC compared with adjacent normal tissue (Atay, 2020).

2.5 Statistical analysis

Using a reference scRNA-seq study (GSE205049), we evaluated median differences in cell type proportion between the PDAC sample and the adjacent normal paired tissue samples from 9 patients using a Wilcoxon Rank Sum Test for 23 cell types. After deconvolution, we created scatter plots to show the correlation across the two RNA-seq tumor samples for each patient and evaluated significant differences in median cell type proportion using a paired Wilcoxon rank sum test for 21 identified cell types across the two paired samples. We used the ESTIMATE algorithm implemented in the tidyestimate package in R to estimate tumor purity (stromal and immune cell signatures) using the normalized RNA-seq data from each study separately. Additionally, we evaluated median differences using a paired Wilcoxon rank sum test across the selected nine genes previously determined to be important in PDAC. We visualized differences in log2 normalized gene expression using violin plot and statistically evaluated the significance of the median difference using a Wilcoxon Rank Sum Test with Holm adjusted p-value with values < 0.05 indicating significance. We also explored the effects of batch correction, normalization, and transformation on median differences. A plot was also generated to show correlations between each of the genes based on Spearman correlation coefficients.

3 Results

3.1 Cell type proportion

We observed that average cell type proportion varied the most between studies (i.e., between samples for each patient) for NK and macrophages (Figures 1A,B). We visually observed across all plots that Macrophages, NK cells, T cells, cycling, and dendritic cells are among the cell types that vary the most. Based on the two single cell reference samples, we observed that NK cells are significantly (adj p-value <0.05) enriched in PDAC tissue compared to adjacent normal tissue while Dendric cells (DC1 and 2) and CD16⁺ monocyte populations are significantly reduced (Figure 2A). Using the Wilcoxon test in our paired samples, we observe the most significant difference in the deconvolution-based cell types are for CD4 (adjusted p-value = 0.00613), endothelial (adjusted p-value = 0.00000172), fibroblasts (adjusted p-value = 0.000161), granulocytes (adjusted p-value = 0.000895), neural (adjusted p-value = 0.000378), NK (adjusted p-value = 0.00224), and monocytes (adjusted p-value = 0.00816; Figure 2B).

Figure 1

Figure 1. (A) Sig 1 and (B) Sig 2 deconvolution cell type proportion estimates for 16 paired samples. Each panel represents a different deconvolution method. Each color represents an individual cell type. Each point represents a patient’s proportion for The Cancer Genome Atlas Program (TCGA) sample (y-axis) and the Mayo sample (x-axis).

Figure 2

Figure 2. (A) Comparison of PDAC (orange) and Adjacent normal (green) cell type proportions. (B) Comparison of Mayo (salmon) vs. TCGA (blue -green) cell type proportions across 21 different cell types. P-values are from the Wilcoxon rank sum test. The scatterplot shows the correlation between TCGA RNA-seq-based deconvolution results and Mayo RNA-seq-based deconvolution. The top row represents more cell types and uses a different single cell reference sample than the bottom row. Each of the three columns represent a different deconvolution method: dtangle, nnis, and qprogwc. Each point is colored according to the cell type, as indicated by the legend on the left and right sides. The primary focus should be on identifying patterns for specific cell types that deviate from the diagonal.

3.2 Tumor infiltration

We estimated tumor infiltration scores in each study separately using normalized RNA-seq data in the ESTIMATE function in R. In summary, this plot (Figure 3) demonstrates the positive relationship between the estimated stromal and immune cell infiltration and the ESTIMATE score, suggesting that tumors with high stromal and immune cells also likely have high infiltration of the tumor. This is expected because a higher presence of non-tumor cells will naturally lower the proportion of tumor cells in the sample. The blue-green (TCGA) and salmon (Mayo) circles highlight specific tumor samples with distinct characteristics in terms of their microenvironment. It is important to note that ESTIMATE scores can only be interpreted relatively, and it cannot be inferred across studies.

Figure 3

Figure 3. Tumor infiltration estimates for Mayo (salmon) and TCGA (blue-green) sample data. The ESTIMATE algorithm from R was used to estimate (A) stromal and (B) immune infiltration in the sample using the normalized gene expression data for each study separately. The tumor infiltration estimate is relative and cannot be compared across studies. (C) Spearman correlation between stromal and immune estimates within and across study. Each significance level is associated to a symbol: p-values (0, 0.001, 0.01, 0.05, 0.1, 1) <=> symbols (“***”, “**”, “*”, “▪”, “ ”).

3.3 Gene expression

Violin plots of gene expression in the paired Mayo and TCGA sample data for the four PDAC specific genes comparing the effect of batch correction, normalization, and transformations are included (Supplementary Figure S1a,b). The plots highlight the importance of adjusting both within sample (normalization) and across sample (batch correction) to avoid bias. When looking at the first four PDAC specific genes, we did not observe significant median expression differences across paired samples. However given our small sample size and low power, these plots suggest there may be potential differences for KRAS (adj. p-value = 0.18) and SMAD4 (adj. p-value = 0.59, Figure 4A). When evaluating the five transcription factor enriched genes, there were no statistically significant results. Visually, CTNNB1(adj. p-value = 0.30) and TCF7 (adj. p-value = 0.56) varied the most between the paired TCGA and Mayo sample data (Figure 4B).

Figure 4

Figure 4. Violin plot of normalized gene expression in transcripts per million (tpm) for Mayo (salmon) and TCGA (blue-green) samples in (A) four selected highly mutated pancreatic cancer genes and (B) five validated highly variably expressed genes. Each gene also has an adjusted p-value for the Wilcoxon comparison between paired samples.

3.4 Concordance and correlation across paired samples

We used the median as a cut point for each gene and calculated one score for each patient separately to determine concordance (Supplementary Table S1). We observed a substantial concordance (kappa = 0.75) for JUN, moderate concordance for KRAS, TP53, SMAD3 and CDKN2A (Kappa = 0.5) and low concordance value for TCF7, SMAD7, SMAD4, CTNNB1(Kappa = 0.25) across the two studies suggesting that sufficiently representative genomic data cannot be collected with one sample. Likewise, the correlation of the gene expression values for the nine selected genes in general vary significantly across the paired samples (Figure 5). In this correlation matrix, each row and column represent a gene and the number in the cell represents the Spearman correlation coefficient with colors ranging from highly positively correlated values (blue) to highly negatively correlated values (red). The coefficients which reached statistical significance ranged from <-0.24 to >0.24 and are shown in the upper triangle. The hierarchical clustering method was used to order the genes in the matrix. We observed moderate positive correlations between TP53/SMAD4 (0.47) among only TCGA samples; SMAD7/CTNNB1 (0.69), KRAS/CTNNB1 (0.36), and SMAD4/CTNNB1 (0.31) using only Mayo samples. We observed a moderate negative correlation between TCF7/CDKN2A (−0.24) within the TCGA samples only. We observed moderate positive correlation across study genes for CDKN2A/CDKN2A (0.52), SMAD4/TCF7 (0.47), and SMAD3/CDKN2A (0.40). Moderate negative correlations were observed for the following genes across studies: SMAD4/CDKN2A (−0.64), TCF7/SMAD3 (−0.46), SMAD7/TP53 (−0.31), and TCF7/CDKN2A (−0.24). These results could suggest that the TME is likely different across the two samples.

Figure 5

Figure 5. Correlation plot showing the Spearman correlation in select gene expression values for TCGA data (blue-green) and Mayo data (salmon). The genes are arranged based on hierarchical clustering. The lower triangle shows all correlation coefficients with higher correlations having higher transparency. The higher triangle shows only the correlation values which have significant p-values. Correlation coefficients with values close to 1 have deeper blue and coefficients with values close to −1 have deeper red color.

4 Discussion

In this study, we observed minimal nonsignificant differences in expression values for selected genes established to have a role in PDAC. This slight difference could be related to the different cell type proportions across the samples that we estimated using deconvolution methods. Studies show that pancreatic tumors are heterogeneous and different regions of the tumor may have distinct genetic and molecular profiles (Lodestijn et al., 2021), so observing a difference across samples would not be surprising. We did not observe significant differences in median log normalized, batch-corrected expression values for our selected genes. However, this may be due to a small sample size and low power in the current study. Collectively, the results suggest there should be additional research investigating how closely a single sample fully captures the heterogeneity of a patient’s tumor and determine how important this might be for treatment planning as genomic results are increasingly used in a clinical setting. Focusing research on how many samples to obtain and strategically deciding where to select a biopsy sample from will be important as single cell and spatial technologies advance.

When evaluating median gene expression across studies, we did not observe any statistically significant differences; however, there are suggestions of a difference. Performing follow-up scRNA-seq studies with a larger sample size could help clarify if there is an actual difference by looking at cell type specific expression. Since 2020, researchers have explored what heterogeneity looks like in PDAC at the single cell level. Key findings include tumor-cancer-associated fibroblast (CAF) interaction (Loveless et al., 2025), defining neural-like progenitor programs in fibroblasts and defining 3 multicellular communities (Hwang et al., 2022), and identifying epithelial and T cells as prognostic factors in PDAC (Du et al., 2024). As we see using the deconvolution and tumor infiltration estimates, there is large variation in the estimated cell type proportions and tumor cells in the sample.

The correlation matrix heatmap provides a comprehensive overview of the pairwise relationships between the expression levels of the given genes. It highlights strong positive and negative correlations that can provide valuable insights into gene co-regulation, functional relationships, and potential regulatory mechanisms. We used hierarchical clustering to order the genes, and that method grouped the genes with sample type intermingled (Mayo vs. TCGA). This could indicate that because cell type proportions vary across samples, there are also potential co-regulation pathways that differ by cell type or based on interacting cells. Hongjing et al. (2025) study observations support this idea by observing complement-secreting CAFs and gap junction-related CAFs locae in spatially distinct regions and demonstrate different co-regulations with tumor cells (Hongjing et al., 2025). Additionally, the difference in correlation structure by sample type could suggest a difference in functional relationships or regulatory mechanisms utilized across the two samples. Cross-sample correlations could potentially suggest a regional relationship between the genes or could just represent random association. Mechanistic studies in the laboratory would need to confirm any of these suggested relationships.

Some of the key limitations in this research include the small sample size, low power, and that each of the paired samples were processed using different sequencing technologies. Any findings should be confirmed with additional studies.

The consistency of RNA-seq results across multiple samples from the same person is a complex issue. Factors that can influence consistency include: 1) biological variation as gene expression fluctuates due to circadian rhythms, physiological state (e.g., stress, illness), and cellular heterogeneity and 2) technical variation as there are many processing steps in the RNA-seq workflow and bioinformatic analysis (Auer and Doerge, 2010). Careful experimental design and quality control are essential to minimize technical variation. While some variability is inherent in biological systems, RNA-seq can provide highly consistent results when proper experimental design and analysis are employed (Su et al., 2014).

Tumor heterogeneity between different sites of neoplasia in a single patient, and among tumors from different patients, is an emerging theme in cancer research (Alizadeh et al., 2015). Heterogeneity confounds researchers’ understanding of tumor evolution and their ability to design effective treatments (Javed et al., 2024; Liu Y. et al., 2024; 2025). Due to the limitations of methods that characterize tumors in ‘bulk’, such as averaging across all tumor clones, even a single RNA-Seq sample, may not fully represent the tumor heterogeneity (Liu Y. et al., 2024; 2025). Ideally, researchers would obtain multiple regions from each tumor to capture spatial heterogeneity (Giannoukakos et al., 2024; Sergeyev et al., 2024; Yaqoob et al., 2024).

While computational deconvolution of different expression components in a sample can distinguish between cells from different lineages, it has limited applicability in samples with low transcriptional diversity (Liu et al., 2025). Even at the single cell level, several factors affect the ability of a single RNA-seq sample to represent tumor heterogeneity (Hwang et al., 2022). Single-cell RNA-Seq is a robust technology that can analyze tens of thousands of cells simultaneously in a cost-effective and efficient manner (Hwang et al., 2022). However, sensitivity for low expressed genes still presents a challenge, and metrics that quantify similarity and difference between samples from the same clonal origin are needed (Hwang et al., 2022; Liu et al., 2025).

The long-term goal of this area of research is to harness the power of genomic analysis to optimize biopsy decisions and ultimately improve patient outcomes. Challenges remain in fully validating genetic and genomic biopsy strategies and supporting their widespread clinical implementation and will be the focus of future work. Attention should focus on standardizing tissue collection/processing protocols, establishing evidence-based thresholds for acting on genomic findings, and integrating molecular and image data into existing clinicopathologic risk assessment frameworks. Research in the future should focus building off the results presented here to get a clearer picture if multiple samples taken at the time of biopsy would help develop more effective treatment plans.

Data availability statement

The datasets analyzed for this study can be found in The Cancer Genome Atlas Program (TCGA-PAAD; portal.gdc.cancer.gov/projects/TCGA-PAAD) and the Gene Expression Omnibus (GEO; https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE229413 and https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE205049). Data generated on the paired samples can be requested from the primary author.

Ethics statement

The studies involving humans were approved by Mayo Clinic institutional review board. The studies were conducted in accordance with the local legislation and institutional requirements. The human samples used in this study were acquired from primarily isolated as part of your previous study for which ethical approval was obtained. Written informed consent for participation was not required from the participants or the participants’ legal guardians/next of kin in accordance with the national legislation and institutional requirements.

Author contributions

RJ: Conceptualization, Formal Analysis, Supervision, Writing – review and editing, Methodology, Investigation, Writing – original draft, Visualization, Funding acquisition. SM: Conceptualization, Visualization, Writing – review and editing, Formal Analysis, Methodology. SA: Writing – review and editing, Methodology. KR: Methodology, Resources, Data curation, Conceptualization, Writing – review and editing. HS: Data curation, Resources, Conceptualization, Methodology, Writing – review and editing.

Funding

The authors declare that financial support was received for the research and/or publication of this article. Partial support from a UMN Data Science Initiative (DSI) and Research Computing SEED grant and NIH grant P30 CA77598 utilizing the Biostatistics Core shared resource of the Masonic Cancer Center, University of Minnesota and by the National Center for Advancing Translational Sciences of the National Institutes of Health Award Number UL1TR002494. Additional partial funding support for this work was provided by National Institutes of Health (NIH) COBRE grant P20GM109024, NIH P50 CA102701, and NIH R25 CA92049. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Acknowledgements

The Authors would like to thank the patients who participated in this research and the TCGA and GEO for providing the publicly accessible data used in this study.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The authors declare that no Generative AI was used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fgene.2025.1662924/full#supplementary-material

References

Alizadeh, A. A., Aranda, V., Bardelli, A., Blanpain, C., Bock, C., Borowski, C., et al. (2015). Toward understanding and exploiting tumor heterogeneity. Nat. Med. 21 (8), 846–853. doi:10.1038/nm.3915

PubMed Abstract | CrossRef Full Text | Google Scholar

Atay, S. (2020). Integrated transcriptome meta-analysis of pancreatic ductal adenocarcinoma and matched adjacent pancreatic tissues. PeerJ 8, e10141. doi:10.7717/PEERJ.10141

PubMed Abstract | CrossRef Full Text | Google Scholar

Auer, P. L., and Doerge, R. W. (2010). Statistical design and analysis of RNA sequencing data. Genetics 185 (2), 405–416. doi:10.1534/genetics.110.114983

PubMed Abstract | CrossRef Full Text | Google Scholar

Baller, J., Kono, T., Herman, A., and Zhang, Y. (2019). ChURP: a lightweight CLI framework to enable novice users to analyze sequencing datasets in parallel. ACM Int. Conf. Proceeding Ser., 1–5. doi:10.1145/3332186.3333156

CrossRef Full Text | Google Scholar

Binkowski, B., Klamer, Z., Gao, C., Staal, B., Repesh, A., Tran, H.-L., et al. (2024). Multiplexed glycan immunofluorescence identification of pancreatic cancer cell subpopulations in both tumor and blood samples. BioRxiv Prepr. Serv. Biol., 2024.08.22.609143. doi:10.1101/2024.08.22.609143

PubMed Abstract | CrossRef Full Text | Google Scholar

Binkowski, B., Klamer, Z., Gao, C., Staal, B., Repesh, A., Tran, H.-L., et al. (2025). Multiplexed glycan immunofluorescence identification of pancreatic cancer cell subpopulations in both tumor and blood samples. Sci. Adv. 11 (10), eadt0029. doi:10.1126/SCIADV.ADT0029

PubMed Abstract | CrossRef Full Text | Google Scholar

Blanco-Heredia, J., Souza, C. A., Trincado, J. L., Gonzalez-Cao, M., Gonçalves-Ribeiro, S., Gil, S. R., et al. (2024). Converging and evolving immuno-genomic routes toward immune escape in breast cancer. Nat. Commun. 15 (1), 1302. doi:10.1038/S41467-024-45292-1

PubMed Abstract | CrossRef Full Text | Google Scholar

Butera, F., Sero, J. E., Dent, L. G., and Bakal, C. (2024). Actin networks modulate heterogeneous NF-κB dynamics in response to TNFα. ELife 13, e86042. doi:10.7554/ELIFE.86042

PubMed Abstract | CrossRef Full Text | Google Scholar

Cai, J., Chen, H., Lu, M., Zhang, Y., Lu, B., You, L., et al. (2021). Advances in the epidemiology of pancreatic cancer: trends, risk factors, screening, and prognosis. Cancer Lett. 520, 1–11. doi:10.1016/j.canlet.2021.06.027

PubMed Abstract | CrossRef Full Text | Google Scholar

Du, J., Zhao, Y., Dong, J., Li, P., Hu, Y., Fan, H., et al. (2024). Single-cell transcriptomics reveal the prognostic roles of epithelial and T cells and DNA methylation-based prognostic models in pancreatic cancer. Clin. Epigenet 16, 188. doi:10.1186/s13148-024-01800-0

PubMed Abstract | CrossRef Full Text | Google Scholar

Giannoukakos, S., D’Ambrosi, S., Koppers-Lalic, D., Gómez-Martín, C., Fernandez, A., and Hackenberg, M. (2024). Assessing the complementary information from an increased number of biologically relevant features in liquid biopsy-derived RNA-seq data. Heliyon 10 (6), e27360. doi:10.1016/j.heliyon.2024.e27360

PubMed Abstract | CrossRef Full Text | Google Scholar

Grünwald, B. T., Devisme, A., Andrieux, G., Vyas, F., Aliar, K., McCloskey, C. W., et al. (2021). Spatially confined sub-tumor microenvironments in pancreatic cancer. Cell 184 (22), 5577–5592.e18. doi:10.1016/J.CELL.2021.09.022

PubMed Abstract | CrossRef Full Text | Google Scholar

Hongjing, Ai, Nie, R., and Wang, X. (2025). Pathway enrichment-based unsupervised learning identifies novel subtypes of cancer-associated fibroblasts in pancreatic ductal adenocarcinoma. Interdiscip. Sci. 17 (2), 477–495. doi:10.1007/s12539-025-00705-7

PubMed Abstract | CrossRef Full Text | Google Scholar

Hu, H. F., Ye, Z., Qin, Y., Xu, X., wu, Yu, jun, X., et al. (2021). Mutations in key driver genes of pancreatic cancer: molecularly targeted therapies and other clinical implications. Acta Pharmacol. Sin. 42 (11), 1725–1741. doi:10.1038/S41401-020-00584-2

PubMed Abstract | CrossRef Full Text | Google Scholar

Hu, J. X., Lin, Y. Y., Zhao, C. F., Chen, W. B., Liu, Q. C., Li, Q. W., et al. (2021). Pancreatic cancer: a review of epidemiology, trend, and risk factors. World J. Gastroenterology 27 (27), 4298–4321. doi:10.3748/wjg.v27.i27.4298

PubMed Abstract | CrossRef Full Text | Google Scholar

Hwang, W. L., Jagadeesh, K. A., Guo, J. A., Hoffman, H. I., Yadollahpour, P., Reeves, J. W., et al. (2022). Single-nucleus and spatial transcriptome profiling of pancreatic cancer identifies multicellular dynamics associated with neoadjuvant treatment. Nat. Genet. 54 (8), 1178–1191. doi:10.1038/S41588-022-01134-8

PubMed Abstract | CrossRef Full Text | Google Scholar

Javed, S., Qureshi, T. A., Wang, L., Azab, L., Gaddam, S., Pandol, S. J., et al. (2024). An insight to PDAC tumor heterogeneity across pancreatic subregions using computed tomography images. Front. Oncol. 14, 1378691. doi:10.3389/fonc.2024.1378691

PubMed Abstract | CrossRef Full Text | Google Scholar

Karmakar, S., Rauth, S., Nallasamy, P., Perumal, N., Nimmakayala, R. K., Leon, F., et al. (2020). RNA polymerase II-Associated factor 1 regulates stem cell features of pancreatic cancer cells, independently of the PAF1 complex, via interactions with PHF5A and DDX3. Gastroenterology 159 (5), 1898–1915.e6. doi:10.1053/J.GASTRO.2020.07.053

PubMed Abstract | CrossRef Full Text | Google Scholar

Kim, D., Paggi, J. M., Park, C., Bennett, C., and Salzberg, S. L. (2019). Graph-based genome alignment and genotyping with HISAT2 and HISAT-genotype. Nat. Biotechnol. 37 (8), 907–915. doi:10.1038/s41587-019-0201-4

PubMed Abstract | CrossRef Full Text | Google Scholar

Li, Z., Zhang, X., Sun, C., Li, Z., Fei, H., and Zhao, D. (2024). Global, regional, and national burdens of early onset pancreatic cancer in adolescents and adults aged 15-49 years from 1990 to 2019 based on the global burden of disease study 2019: a cross-sectional study. Int. J. Surg. Lond. Engl. 110 (4), 1929–1940. doi:10.1097/JS9.0000000000001054

PubMed Abstract | CrossRef Full Text | Google Scholar

Liao, Y., Smyth, G. K., and Shi, W. (2014). featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinforma. Oxf. Engl. 30 (7), 923–930. doi:10.1093/BIOINFORMATICS/BTT656

PubMed Abstract | CrossRef Full Text | Google Scholar

Liu, R., Qiu, M., Deng, X., Zhang, M., Gao, Z., Wang, Y., et al. (2024). Erianin inhibits the progression of pancreatic cancer by directly targeting AKT and ASK1. Cancer Cell Int. 24 (1), 348. doi:10.1186/S12935-024-03533-9

PubMed Abstract | CrossRef Full Text | Google Scholar

Liu, Y., Carbonetto, P., Willwerscheid, J., Oakes, S. A., Macleod, K. F., and Stephens, M. (2024). Dissecting tumor transcriptional heterogeneity from single-cell RNA-seq data by generalized binary covariance decomposition. BioRxiv Prepr. Serv. Biol., 2023.08.15.553436. doi:10.1101/2023.08.15.553436

PubMed Abstract | CrossRef Full Text | Google Scholar

Liu, Y., Carbonetto, P., Willwerscheid, J., Oakes, S. A., Macleod, K. F., and Stephens, M. (2025). Dissecting tumor transcriptional heterogeneity from single-cell RNA-seq data by generalized binary covariance decomposition. Nat. Genet. 57 (1), 263–273. doi:10.1038/S41588-024-01997-Z

PubMed Abstract | CrossRef Full Text | Google Scholar

Lodestijn, S. C., Miedema, D. M., Lenos, K. J., Nijman, L. E., Belt, S. C., El Makrini, K., et al. (2021). Marker-free lineage tracing reveals an environment-instructed clonogenic hierarchy in pancreatic cancer. Cell Rep. 37 (3), 109852. doi:10.1016/J.CELREP.2021.109852

PubMed Abstract | CrossRef Full Text | Google Scholar

Loveless, I. M., Kemp, S. B., Hartway, K. M., Mitchell, J. T., Wu, Y., Zwernik, S. D., et al. (2025). Human pancreatic cancer single-cell atlas reveals association of CXCL10+ fibroblasts and basal subtype tumor cells. Clin. Cancer Res. 31 (4), 756–772. doi:10.1158/1078-0432.CCR-24-2183

PubMed Abstract | CrossRef Full Text | Google Scholar

Mazur, P. K., Grüner, B. M., Nakhai, H., Sipos, B., Zimber-Strobl, U., Strobl, L. J., et al. (2010). Identification of epidermal Pdx1 expression discloses different roles of Notch1 and Notch2 in murine Kras(G12D)-induced skin carcinogenesis in vivo. PloS One 5 (10), e13578. doi:10.1371/JOURNAL.PONE.0013578

PubMed Abstract | CrossRef Full Text | Google Scholar

Nair, S. T., Abhi, C., Kamalasanan, K., Pavithran, K., Unni, A. R., Sithara, M. S., et al. (2024). Pathophysiology-driven approaches for overcoming nanomedicine resistance in pancreatic cancer. Mol. Pharm. 21 (12), 5960–5988. doi:10.1021/acs.molpharmaceut.4c00801

PubMed Abstract | CrossRef Full Text | Google Scholar

Oketch, D. J. A., Giulietti, M., and Piva, F. (2024). Copy number variations in pancreatic cancer: from biological significance to clinical utility. Int. J. Mol. Sci. 25 (1), 391. doi:10.3390/ijms25010391

PubMed Abstract | CrossRef Full Text | Google Scholar

Padwal, M. K., Parghane, R. V., Chakraborty, A., Ujaoney, A. K., Anaganti, N., Basu, S., et al. (2024). Developing a peripheral blood RNA-seq based NETseq ensemble classifier: a potential novel tool for non-invasive detection and treatment response assessment in neuroendocrine tumor patients receiving 177Lu-DOTATATE PRRT. J. Neuroendocrinol. 37, e13462. doi:10.1111/JNE.13462

PubMed Abstract | CrossRef Full Text | Google Scholar

Pfister, S., Kuettel, V., and Ferrero, E. (2024). Granulator: rapid benchmarking of methods for *in silico* deconvolution of bulk RNA-seq data. R package version 1.14.0.

Google Scholar

Rawla, P., Sunkara, T., and Gaduputi, V. (2019). Epidemiology of pancreatic cancer: global trends, etiology and risk factors. World J. Oncol. 10 (1), 10–27. doi:10.14740/wjon1166

PubMed Abstract | CrossRef Full Text | Google Scholar

Sergeyev, O., Bezuglov, V., Soloveva, N., Smigulina, L., Denisova, T., Dikov, Y., et al. (2024). Intraindividual variability of semen quality, proteome, and sncRNA profiles in a healthy cohort of young adults. Andrology 13, 840–859. doi:10.1111/andr.13739

PubMed Abstract | CrossRef Full Text | Google Scholar

Su, Z., Łabaj, P. P., Li, S., Thierry-Mieg, J., Thierry-Mieg, D., Shi, W., et al. (2014). A comprehensive assessment of RNA-seq accuracy, reproducibility and information content by the sequencing quality control consortium. Nat. Biotechnol. 32 (9), 903–914. doi:10.1038/nbt.2957

PubMed Abstract | CrossRef Full Text | Google Scholar

Yaqoob, A., Verma, N. K., Aziz, R. M., and Shah, M. A. (2024). RNA-seq analysis for breast cancer detection: a study on paired tissue samples using hybrid optimization and deep learning techniques. J. Cancer Res. Clin. Oncol. 150 (10), 455. doi:10.1007/s00432-024-05968-z

PubMed Abstract | CrossRef Full Text | Google Scholar

Ying, H., Kimmelman, A. C., Bardeesy, N., Kalluri, R., Maitra, A., and DePinho, R. A. (2025). Genetics and biology of pancreatic ductal adenocarcinoma. Genes and Dev. 39 (1–2), 36–63. doi:10.1101/gad.351863.124

PubMed Abstract | CrossRef Full Text | Google Scholar