Biology subjects

Pancaldi, V.

Publications and source records attributed to Pancaldi, V..

At least 19 recordsLinked to original sources

bioRxiv2026

A multiscale pharmacokinetic-Boolean modelling framework for ibrutinib resistance in a Chronic Lymphocytic Leukaemia case study

Ibrutinib, a covalent Bruton's tyrosine kinase (BTK) inhibitor, achieves durable remissions in chronic lymphocytic leukaemia (CLL), and yet a substantial proportion of patients relapse. A fundamental question is how the quantitative dynamics of drug exposure relate to the regulatory programmes and cell fate decisions that characterise response and relapse. Addressing it requires a multiscale framework that bridges the quantitative pharmacology of drug exposure with the molecular mechanisms encoded in the gene regulatory network, a coupling that has remained largely unexplored in the context of targeted therapy resistance. Here we present a multi-scale computational framework that addresses this question by coupling continuous patient-specific pharmacokinetic profiles with discrete Boolean dynamics, bridging two modelling paradigms that have so far operated in isolation. Applied to a longitudinal single-cell dataset from one CLL patient treated with ibrutinib across three clinical timepoints (pre-treatment, response, relapse), the framework integrates gene regulatory network inference, cell fate trajectory analysis, and a stochastic Boolean model, whose dynamics is driven by the patient-specific monthly drug exposure. AP-1 and integrated stress response regulons were specific to relapse, and the coupled model reproduced the clinical course. Its perturbation screens identified co-inhibition of BTK with BCL2L1 or NFKB1 as synergistic, in line with the rationale for dual targeting of BTK and the BCL2 family in treatment-naive CLL. These results show how the framework turns single-patient data into mechanistic, testable hypotheses on resistance and combination therapy, which can be evaluated in larger cohorts and in experimental models.

systems biology↗

bioRxiv2026

A PTBP1-CDC42 splicing axis regulates leukemia growth and venetoclax sensitivity in acute myeloid leukemia

Acute myeloid leukemia (AML) is an aggressive blood cancer characterized by high rates of relapse and poor outcomes, especially in elderly or unfit patients, who cannot tolerate intensive chemotherapy. While the BCL2 inhibitor venetoclax has improved initial responses in this high-risk population, relapses remain nearly universal, highlighting the need for novel therapeutic strategies. Here, we identify the RNA-binding protein PTBP1 as a critical dependency in AML. PTBP1 depletion impairs leukemic growth in vitro and in vivo, and is associated with widespread splicing alterations and global disruption of protein synthesis. Integrative transcriptomic and iCLIP analyses reveal that PTBP1 orchestrates a splicing program centered on Rho GTPase signaling, with CDC42 as a key downstream effector. Mechanistically, PTBP1 loss triggers a splicing switch from CDC42-v1 to CDC42-v2, leading to reduced GTPase activity and impaired protein synthesis. Pharmacological inhibition of CDC42 selectively induces cytotoxicity in AML cells, while sparing healthy hematopoietic cells. Importantly, CDC42 inhibition markedly enhances venetoclax anti-leukemic efficacy. These findings establish PTBP1 as a critical regulator of AML cell fitness and identify a clinically actionable therapeutic combination that exploits AML dependency on PTBP1-CDC42 signaling to enhance the efficacy of venetoclax-based regimens.

cancer biology↗

bioRxiv2026

CellTFusion: a transcriptional regulatory network framework for the identification of functional multicellular states from bulk RNA-seq data

Bulk RNA-seq remains the most accessible transcriptomic platform for tumor microenvironment (TME) characterization, yet existing computational approaches treat cell type abundance, pathway activity, and transcription factor (TF) activity as independent sources of information, missing the coordinated regulatory programs that define functional multicellular states. Here we introduce CellTFusion, a framework that integrates cell type deconvolution with transcriptional regulatory network analysis from bulk RNA-seq data to identify functional multicellular groups. CellTFusion produces a mixture representation of the TME in which each patient is described as a weighted combination of states, each capturing a distinct coordinated hallmark program. Applied to melanoma and bladder cancer cohorts, CellTFusion identified recurrent TME programs with opposing associations with immunotherapy responses that only emerged through joint multivariate modeling, and demonstrated superior cross-cohort transferability compared to established TME characterization tools.

bioinformatics↗

bioRxiv2026

Specific dendritic cells spatial organization is associated to ICB Response in Non Small Cell Lung Cancer (NSCLC)

Dendritic cells (DCs) are central orchestrators of antitumor immunity. Several DC subsets--including conventional type 1 (cDC1), conventional type 2 (cDC2), plasmacytoid DCs (pDCs), and mature DC populations--play distinct roles in immune surveillance, tumor control, immunotherapy response and prognosis. Recent findings suggest that cDC1 are spatially closed to CD8 T-cell and contribute to tertiary lymphoid structure formation in lung cancer. However, how other DC subsets interact with cDC1 to shape the tumor microenvironment (TME) remains largely unknown. Here, we analyzed the spatial distribution of major DC subsets, including cDC1, cDC2, mature DC and pDC, together with CD8 T cells in a cohort of anti-PD1-treated NSCLC patients and we deciphered the corresponding immune microenvironment behavior by paired transcriptomic analysis. We found that, while other DC subsets populated the stroma, cDC2 were localized both in the stroma and in tumor nests. Moreover, unlike other DC subsets, cDC2 abundancy did not affect ICB response both at transcriptomic and in situ analysis. We described spatial organization of DCs in megaclusters characterized by distinct proportions of DC subsets. Patients enriched in megaclusters involving variable proportion of pDC, cDC1 and mature DC, exhibited pro-inflammatory transcriptomic programs while those enriched in cDC2-based megaclusters showed limited immune activation features. Globally, DC in lung cancer were structured around three distinct DC spatial patterns, namely cDC1-driven, cDC2-driven and DC-Scattered, each defined by unique compositions of DC megaclusters, immune features and pathways activation profiles. Among them, the cDC1-driven pattern was associated to prolonged anti-PD1 response in two independent cohorts.

cancer biology↗

bioRxiv2026

A new pipeline for cross-validation fold-aware machine learning prediction of clinical outcomes addresses hidden data-leakage in omics based 'predictors'.

MotivationMachine learning (ML) approaches are increasingly applied to high-dimensional biological data in which features are often dataset-dependent. In many omics workflows, features are computed using information derived from the entire dataset, such as correlations between variables, clustering structures, or enrichment scores. We refer to these as global dataset features, defined as features whose computation depends on properties of the full dataset. In such cases, standard validation strategies can fail, especially when evaluating on independent datasets, due to information leakage that leads to overly optimistic performance estimates. ResultsTo address this challenge, we present pipeML, a flexible and modular machine learning framework designed to support leakage-free model training through custom cross-validation (CV) fold construction. pipeML enables users to recompute global dataset features independently within each CV fold, ensuring strict separation between training and test data, while preserving compatibility with a wide range of ML algorithms for both classification and survival tasks. Using real-world biological datasets, we demonstrate that pipeML enables leakage-free model evaluation when global dataset features are used. We argue that overestimation of model performance during CV can lead to overoptimistic expectations for validation on independent datasets. By explicitly addressing data leakage and offering a transparent, modular workflow, pipeML provides a robust solution for developing and validating ML models in complex biological settings. AvailabilityThe pipeML R package as well as a tutorial are available at https://github.com/VeraPancaldiLab/pipeML Contactvera.pancaldi@inserm.fr or marcelo.hurtado@inserm.fr Supplementary informationAvailable at Bioinformatics online.

bioinformatics↗

bioRxiv2025

PhysiGym: bridging the gap between the Gymnasium reinforcement learning application interface and the PhysiCell agent-based model software

This paper presents PhysiGym, a framework that integrates agent-based biological simulation within standardized reinforcement learning environments. By integrating the agent-based modeling framework PhysiCell with the Gymnasium API, we provide a flexible tool for exploring reinforcement learning strategies to control insilico biological processes. We demonstrate PhysiGyms potential with a case study where a deep reinforcement learning algorithm guides a tumor microenvironment model toward an anti-tumoral state, ultimately achieving tumour elimination. Our results highlight PhysiGyms flexibility for AI-driven biological control and optimization of dynamic treatment regimes.

bioinformatics↗

bioRxiv2025

Targeting human M2 macrophages with antibodies in optimized 3D tumor spheroids

Tumor microenvironment (TME) constituents, including tumor-associated macrophages (TAMs), are now well known to have a significant impact on tumor development. TAMs can be predominant cells in the TME, able to promote cancer cell proliferation, resistance to treatments and immunosuppression. Inactivating these TAMs in tumors by their depletion or their repolarization into a specific anti-tumor, inflammatory phenotype constitutes a significant challenge in immuno-oncology. Several tools and methods to target TAMs have been proposed but they often show low specificity for pro-tumoral macrophages and target myeloid cells too broadly. This research area is therefore in full expansion and warrants the development of appropriate study models. Given the current global effort to reduce the use of in vivo approaches, developing in vitro models that mimic the behavior of TAMs in the TME has become a priority. In this study, we focused on targeting pro-tumoral TAMs in a simple in vitro model recapitulating cancer cell proliferation and TAMs specific phenotype. We developed a 3D model consisting of cancer cells and pro-tumoral macrophages, in which these macrophages promote tumor cell proliferation, maintaining an immunosuppressive phenotype. We showed that, in this model, macrophages can be easily targeted and killed with specific antibodies even in the central regions of the spheroid. This model could be a first in vitro approach to screen new TAM depletion or depolarization tools before application in in vivo models.

cancer biology↗

bioRxiv2025

multideconv - Integrative pipeline for cell type deconvolution from bulk RNAseq using first and second generation methods

The number of computational methods for cell type deconvolution from bulk RNA-seq data has been increasing in the last years, but their high feature complexity and variability of results across methods and signatures limit their utility and effectiveness for patient stratification. Applying multiple combinations of deconvolution methods and signatures often results in hundreds of redundant or contradictory cell type features describing the composition of complex tumour samples. Benchmarking efforts are inherently limited by the lack of bias-free ground truth, often yielding inconsistent results or no consensus. To address these limitations, we present multideconv, an R package that reduces dimensionality and eliminates redundancy in deconvolution results, through unsupervised filtering and iterative correlation analyses. Built on top of existing frameworks, multideconv harmonizes outputs across methods to identify robust cell type proportion estimates and mitigate signature-driven heterogeneity. We benchmarked multideconv against two existing methods that provide similar functions and found it to yield more accurate estimations of cell type proportions based on virtual bulk reconstruction from single-cell expression datasets. We also increase computational efficiency by providing a meta-cell aggregation of the single-cell datasets, showing it preserves the samples complexity. Despite our focus on tumour samples in the context of immuno-oncology, the tool is flexible and can be adapted to infer mixed sample composition from bulk RNAseq datasets. The multideconv R package and tutorials are available at https://github.com/VeraPancaldiLab/multideconv. The code to reproduce the analysis and figures is available on github at https://github.com/VeraPancaldiLab/multideconv_paper. Contact: marcelo.hurtado@inserm.fr or vera.pancaldi@inserm.fr

bioinformatics↗

bioRxiv2025

Gut microbial ecosystems differ across metabolic and obesity phenotypes

Obesity is a heterogeneous condition comprising a continuum of phenotypes with various metabolic and inflammatory profiles. Metabolically healthy obesity (MHO) identifies individuals with obesity but a relatively preserved metabolic state. However, the criteria defining MHO remain inconsistent, and little is known about the gut microbiome (GM) features underlying this intermediate phenotype. Here, we aim to describe microbial structures contributing to metabolic health and disease. To do so, we analyzed the GM of 959 individuals classified as metabolically healthy non-obese (MHNO), MHO, metabolically unhealthy non-obese (MUNO), and metabolically unhealthy obese (MUO), using stool shotgun metagenomics. MHO subjects display intermediate anthropometric and biochemical profiles, with a GM composition and diversity in an in-between state among MHNO and MUO individuals. Network science analyses reveal that metabolic health, rather than obesity, drives microbial connectivity: MHNO and MHO individuals harbor more robust and functionally cohesive microbial networks, whose most influential nodes are focused toward SCFA production. In contrast, MUO and MUNO communities exhibit a dysbiotic state with reduced connectivity and increased influence of low-abundance, ectopic and potentially pro-inflammatory species resulting in a damaged, unstable microbial community network. These findings suggest that metabolic disorders disrupt microbial ecology beyond compositional shifts, emphasizing the need for systems-level approaches. Our findings show differences in microbial connectivity and association patterns across metabolic and obesity phenotypes, shedding light on how distinct microbial structures may contribute to metabolic health and disease.

bioinformatics↗

bioRxiv2025

Time-series RNA-Seq and data-driven network inference unveil dynamics of cell activation, survival and crosstalk in Chronic Lymphocytic Leukaemia in vitro models

How do cancer cells respond to their environment, and what are the key regulators behind their behaviour? While immune cell reprogramming in the tumour microenvironment (TME) has been extensively studied, the dynamic regulatory changes within cancer cells in response to interactions with immune cells remain poorly understood. In Chronic Lymphocytic Leukaemia (CLL), this knowledge gap limits our ability to fully grasp the disease progression and to design effective, personalised interventions. To tackle this, we combine time-series transcriptomics with data-driven gene regulatory network (GRN) inference to uncover the temporal regulatory mechanisms driving CLL cell behaviour within a reconstituted in vitro TME. Using cultures of peripheral blood from CLL patients or of purified patient-derived CLL cells, we profile gene expression across five time points spanning 14 days under these experimental conditions. By inferring GRNs from transcription factor activity, we capture patient-specific and temporally resolved regulatory interactions that highlight how immune signals shape cancer cell phenotypic changes. Our network analysis reveals distinct gene modules associated with critical processes such as cytokine signalling, metabolic reprogramming and differentiation, hallmarks of immune-cancer cell interaction. Intriguingly, we found that while the presence of immune cells in the environment significantly alters CLL cell activation, their survival trajectories are predominantly governed by intrinsic features. This study not only offers mechanistic insights into how immune cell presence influences CLL cell fate but also presents a robust computational framework for integrating time-series transcriptomics with GRN inference, which can then be used to study the long-term behaviour of the CLL cells through dynamical modelling.

systems biology↗

bioRxiv2025

Unraveling the network signatures of oncogenicity in virus-human protein-protein interactions

Climate change, increasing urbanization and global human mobility have increased the risk of emergent infectious diseases and the risk for their pandemic potential. Consequently, there is a pressing need for methods that can provide rapid insights about the potential enduring effects on the human hosts. In-silico methods can be more suitable than in-vitro and in-vivo methods to identify effects potentially manifesting after a long time, under the hypothesis that the underlying causes can be ascribed to perturbed biomolecular processes. Here we focus on oncogenicity, i.e., the risk of developing cancer subsequent to a viral infection, a phenomenon estimated to account for approximately 15% of global cancer prevalence. We characterize viruses in terms of their oncogenic potential by analyzing a multilayer representation of protein-protein interaction networks reconstructed from the human interactome. Our frame-work facilitates classification and allows us to identify sub-sets of proteins in-volved in oncogenesis, highlighting the mechanisms underpinning viral oncogenicity, including factors like chromatin structure. Our framework serves as a foundational resource for medical research, paving the way for the study of other potential long-term effects associated with emerging viruses.

systems biology↗

bioRxiv2025

Trogocytosis-mediated transfer of FOLR2 from Nurse-like cells to CLL cells is linked to their activation and proliferation

Communication with the lymphoid microenvironment is crucial for survival and proliferation of neoplastic B cells in chronic lymphocytic leukemia (CLL). In this study, we examined nurse-like cells (NLCs), CLL-specific macrophages, strongly implicated in CLL pathogenesis, and their interaction with leukemic cells. Using primary patient cells, we demonstrated that NLCs express high levels of folate receptor beta (FOLR2), which correlates with increased survival of cancer cells in vitro. Furthermore, we discovered that CLL cells acquire functional FOLR2 from NLCs via trogocytosis, enhancing their folate uptake. By mimicking the CLL microenvironment with soluble factors, CD40L and IL-15, we observed a strong NLC-dependent CLL cell activation and proliferation. Moreover, we linked this phenomenon with trogocytosis, by demonstrating that FOLR2+ CLL cells are the predominant population of actively cycling cancer cells. By multiplex immunofluorescence analysis of CLL patient lymph nodes, we confirmed the presence of FOLR2 NLCs, and observed their enrichment in more aggressive and proliferative CLL cases. Finally, we detected FOLR2 cancer cells, providing evidence of trogocytosis in situ. Taken together, we propose FOLR2 as a novel marker of protective NLCs, and highlight the importance of trogocytosis in improved CLL cell adaptation, including NLC-mediated activation and proliferation. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=58 SRC="FIGDIR/small/630890v3_ufig1.gif" ALT="Figure 1"> View larger version (22K): org.highwire.dtl.DTLVardef@1cde6d6org.highwire.dtl.DTLVardef@ee0f67org.highwire.dtl.DTLVardef@1355cb3org.highwire.dtl.DTLVardef@590b29_HPS_FORMAT_FIGEXP M_FIG O_FLOATNOScheme 1.C_FLOATNO Schematic representation of trogocytosis-mediated acquisition of NLC-derived FOLR2 by CLL cells, and its impact on promoting cancer cell adaptability to folate deprivation. 1) CLL cells enter the lymph node (LN) and engage in contact with FOLR2+ NLCs. 2) The presence of T cell- and stromal cell-derived cytokines, CD40L and IL-15 promotes the activation and trogocytosis-mediated acquisition of FOLR2 by CLL cells. 3) FOLR2 trogocytic CLL cells have an advantage in folate acquisition and show increased proliferation compared to the remaining population of cancer cells C_FIG SUMMARYO_LICLL cells acquire functional FOLR2 from NLCs via trogocytosis, a process that is closely linked with enhanced CLL cell activation and proliferation in vitro C_LIO_LIIncreased Frequency of FOLR2 expressing macrophages in more aggressive and proliferative cases of CLL. C_LIO_LIFOLR2 is a marker of M2-like protective NLCs in vitro C_LI

cancer biology↗

bioRxiv2024

Unveiling epigenetic regulatory elements associated with breast cancer development

Breast cancer is the most common cancer in women and the 2nd most common cancer worldwide, yearly impacting over 2 million females and causing 650 thousand deaths. It has been widely studied, but its epigenetic variation is not entirely unveiled. We aimed to identify epigenetic mechanisms impacting the expression of breast cancer related genes to detect new potential biomarkers and therapeutic targets. We considered The Cancer Genome Atlas database with over 800 samples and several omics datasets such as mRNA, miRNA, DNA methylation, which we used to select 2701 features that were statistically significant to differ between cancer and control samples using the Monte Carlo Feature Selection and Interdependency Discovery algorithm, from an initial total of 417,486. Their biological impact on cancerogenesis was confirmed using: statistical analysis, natural language processing, linear and machine learning models as well as: transcription factors identification, drugs and 3D chromatin structure analyses. Classification of cancer vs control samples on the selected features returned high classification weighted Accuracy from 0.91 to 0.98 depending on feature-type: mRNA, miRNA, DNA methylation, and classification algorithm. In general, cancer samples showed lower expression of differentially expressed genes and increased {beta}-values of differentially methylated sites. We identified mRNAs whose expression is well explained by miRNA expression and differentially methylated sites {beta}-values. We recognized differentially methylated sites possibly affecting NRF1 and MXI1 transcription factors binding, causing a disturbance in NKAPL and PITX1 expression, respectively. Our 3D models showed more loosely packed chromatin in cancer. This study successfully points out numerous possible regulatory dependencies.

cancer biology↗

bioRxiv2023

Chromatin network rearrangements in differentiation and cancer reveal connections between evolutionary processes and gene regulation

Cancer cells are highly plastic, allowing them to adapt to changing conditions. Genes related to basic cellular processes evolved in ancient species, while more specialized genes appeared later with multicellularity (metazoan genes) or even after mammals evolved. Transcriptomic analyses have shown that ancient genes are up-regulated in cancer, while metazoan-origin genes are inactivated. Despite the importance of these observations, the underlying mechanisms remain unexplored. Here, we study local and global epigenomic mechanisms that may regulate genes from specific evolutionary periods. Using evolutionary gene age data, we characterize the epigenomic landscape, gene expression regulation, and chromatin organization in three cell types: human embryonic stem cells, normal B-cells, and primary cells from Chronic Lymphocytic Leukemia, a B-cell malignancy. We identify topological changes in chromatin organization during differentiation observing patterns in Polycomb repression and RNA Polymerase II pausing, which are reversed during oncogenesis. Beyond the non-random organization of genes and chromatin features in the 3D epigenome, we suggest that these patterns lead to preferential interactions among ancient, intermediate, and recent genes, mediated by RNA Polymerase II, Polycomb, and the lamina, respectively. Our findings shed light on gene regulation according to evolutionary age and suggest this organization changes across differentiation and oncogenesis.

cancer biology↗

bioRxiv2023

mosna reveals different types of cellular interactionspredictive of response to immunotherapies in cancer

Spatially resolved omics enable the discovery of tissue organization of biological or clinical importance. Despite the existence of several methods, performing a rational analysis including multiple algorithms while integrating different conditions such as clinical data is still not trivial. To make such investigations more accessible, we developed mosna, a Python package to analyze spatial omics data in integration with clinical or biological data, providing insight on cell interaction patterns or tissue architecture. mosna is compatible with all spatial omics techniques, it leverages tysserand to build accurate spatial networks, and is compatible with Squidpy. It proposes an analysis pipeline, in which increasingly complex features computed at each step with either the mosna- algorithms or others can be explored in integration with clinical data. The approach produces easy-to-use descriptive statistics and data visualization, while seamlessly training machine learning models and identifying variables with the most predictive power. mosna can take as input any dataset produced by spatial omics methods, including subcellular resolved transcriptomics (MERFISH, seqFISH, Xenium) and proteomics (CODEX, MIBI-TOF, low-plex immuno-fluorescence), as well as spot-based spatial transcriptomics (10x Visium, Slide-seq, Stereo-seq). Integration with experimental metadata or clinical data is adapted to binary conditions, such as biological treatments or response status of patients, and to survival data. We demonstrate the proposed analysis pipeline on two spatially resolved proteomic datasets and a spatial transcriptomics dataset containing either binary response to immunotherapy or survival data, and we assess the performance of the proposed niche discovering method in a manually annotated spatial transcriptomic dataset. mosna identifies features describing cellular composition and spatial patterns that can provide biological insight regarding factors that affect response to immunotherapies or survival. Availability and implementationmosna is made publicly available to the community, together with relevant documentation at https://mosna-documentation.readthedocs.io/en/latest/index.html and tutorials implemented as Jupyter notebooks to reproduce the result at https://github.com/AlexCoul/mosna

bioinformatics↗

bioRxiv2021

An Agent-Based Model of monocyte differentiation into Tumour-Associated Macrophages in Chronic Lymphocytic Leukemia

Monocyte-derived macrophages help maintain tissue homeostasis and defend the organism against pathogens. In tumors, recent studies have uncovered complex macrophage populations, including tumor-associated macrophages, which support tumorigenesis through cancer hallmarks such as immunosuppression, angiogenesis or matrix remodeling. In the case of chronic lymphocytic leukemia, these macrophages are known as nurse-like cells and they protect leukemic cells from spontaneous apoptosis contributing to their chemoresistance. We propose an agent-based model of monocyte differentiation into nurse-like cells upon contact with leukemic B cells in vitro. We performed patient-specific model calibrations using cultures of peripheral blood mononuclear cells from patients. Using our model, we were able to reproduce temporal survival dynamics of cancer cells in a patient-specific manner and to identify patient groups related to distinct macrophage phenotypes. Our results show a potentially important role of phagocytosis in the polarization process of nurse-like cells and in promoting cancer cells enhanced survival.

systems biology↗

bioRxiv2021

Cytidine deaminase protects pancreatic cancer cells from replicative stress and drives resistance to DNA-targeting drugs

Chronic DNA replication stress and genome instability are two hallmarks of cancer that fuel oncogenesis and tumor diversity. Therapeutic approaches aimed to leverage tumor-specific replication stress to intolerable levels or to expose vulnerabilities for synthetic lethality purposes have recently gained momentum, especially for pancreatic cancer, a disease with no cure. However, the current knowledge regarding the molecular mechanisms involved in the replication stress response in pancreatic tumors is limited. Cytidine deaminase (CDA) is involved in the pyrimidine salvage pathway for DNA and RNA synthesis. Loss of CDA induces genomic instability in Bloom Syndrome, and CDA protects tumor cells from chemotherapy with pyrimidine analogs. Here, we show that CDA is overexpressed in genetically unstable pancreatic tumors, associates with a DNA replication signature, and is instrumental for experimental tumor growth. In cancer cells, CDA promotes DNA replication, increases replication fork speed, and controls replication stress and genomic stability levels. CDA expression is predictive of DNA-damaging drug efficacy and targeting CDA relieves resistance to chemotherapy in patients models, both in vitro and in vivo. Our findings shed new light on the mechanisms by which pancreatic cancer cells control replication stress, and highlight targeting of CDA as a potential therapeutic strategy to defeat tumor resistance to treatment.

cancer biology↗

bioRxiv2021

GEM-DeCan: Improving tumor immune microenvironment profiling by the integration of novel gene expression and DNA methylation deconvolution signatures

Quantifying the proportion of the different cell types present in tumor biopsies remains a priority in cancer research. So far, a number of deconvolution methods have emerged for estimating cell composition using reference signatures, either based on gene expression or on DNA methylation from purified cells. These two deconvolution approaches could be complementary to each other, leading to even more performant signatures, in cases where both data types are available. However, the potential relationship between signatures based on gene expression and those based on DNA methylation remains underexplored. Here we present five new deconvolution signature matrices, based RNAseq data or on DNA methylation, which can estimate the proportion of immune cells and cancer cells in a tumour sample. We test these signature matrices on available datasets for in-silico and in-vitro mixtures, peripheral blood, cancer samples from TCGA, and a single-cell melanoma dataset. Cell proportions estimates based on deconvolution performed using our signature matrices, implemented within the EpiDISH framework, show comparable or better correlation with FACS measurements of immune cell-type abundance and with various estimates of cancer sample purity and composition than existing methods. Using publicly available data of 3D chromatin structure in haematopoietic cells, we expanded the list of genes to be included in the RNAseq signature matrices by considering the presence of methylated CpGs in gene promoters or in genomic regions which are in 3D contact with these promoters. Our expanded signature matrices have improved performance compared to our initial RNAseq signature matrix. Finally, we show the value of our signatures in predicting patient response to immune checkpoint inhibitors in three melanoma cancer cohorts, based on bulk tumour sample gene expression. We also provide GEM-DeCan: a snakemake pipeline, able to run an analysis from raw sequencing data to deconvolution based on various gene expression signature matrices, both for bulk RNASeq and DNA methylation data.

bioinformatics↗