Biology subjects

Dey, M.

Publications and source records attributed to Dey, M..

13 recordsLinked to original sources

bioRxiv2026

Agentic Discovery Transfer Enables Cross-Cancer Discovery at Scale

Cancer research generates tens of thousands of studies each year, yet whether a published finding holds in another cancer type or patient population is rarely tested, because doing so is labor-intensive and difficult to scale. We define this task as discovery transfer and present SMART, a multi-agent large language model framework for discovery transfer in single-cell RNA-sequencing (scRNA-seq) studies. Given a source manuscript and a target scRNA-seq dataset, SMART extracts major clinical and mechanistic findings, generates and executes analyses for the testable findings, and reports which findings transfer and suggests follow-up. Applied to 121 studies across six cancer types, SMART identified findings that transfer across cancers, including a lung cancer prognostic signature that we confirmed to be associated with survival in liver cancer, pancreatic cancer and pleural mesothelioma. SMART also generated new mechanistic hypotheses, demonstrating how agentic discovery transfer can extend published findings beyond their original setting.

bioinformatics↗

bioRxiv2026

An Evolutionarily Conserved Ypk1/SGK1 Kinase Pathway Regulates the Unfolded Protein Response by Modulating the Ire1 Protein Abundance

The unfolded protein response (UPR) is a cellular mechanism that maintains protein homeostasis (proteostasis) under conditions of endoplasmic reticulum (ER) stress. The dual kinase/RNase Ire1 is a conserved regulator of the UPR, mediating the unconventional cytosolic splicing of HAC1 mRNA in yeast and XBP1 mRNA in human cells. The resulting spliced HAC1/XBP1 transcript encodes a transcription factor that induces the expression of protein-folding chaperones and stress-responsive genes, thereby restoring proteostasis. In our previous work, we showed that the MAP kinase Slt2 (homolog of human ERKs) contributes to UPR signaling by promoting IRE1 expression through the transcription factor Rlm1 (homolog of human MEF2C). Here, we demonstrate that Hac1 expression is reduced in yeast strains deficient in essential protein kinase Cdc28, Pkc1, Rio2, Tor2, Pkh1, or Ypk1, suggesting that these kinases also serve as UPR regulators. We focused on the kinase Ypk1, the yeast ortholog of human SGK1 (serum/glucocorticoid-regulated kinase 1). We provide genetic and biochemical evidence that Ypk1/SGK1 acts upstream of the Pkc1/PKC{delta} signaling pathway and is required for maintaining IRE1 protein abundance in both yeast and human cells. Collectively, our results identify an evolutionarily conserved Ypk1/SGK1 signaling pathway that regulates the HAC1 and XBP1 mRNA splicing by modulating the Ire1 protein abundance.

genetics↗

bioRxiv2026

Proteo-transcriptomics and morphometrics of teleost cardiac cells define regulatory networks and exercise-induced cardiomyocyte hypertrophy and hyperplasia

Zebrafish and medaka are powerful cardiovascular models, yet cellular and molecular investigations of adult heart cells have been constrained by suboptimal dissociation and characterization methods. To overcome these barriers, we developed a physiological-temperature workflow that generates high-yield, viable single-cell suspensions for FACS, imaging, and low-input molecular profiling. Using transgenic fluorescent reporters, we consistently isolate [~]6,000 cardiomyocytes per adult zebrafish ventricle and [~]12,000 from medaka, preserving cellular, structural, and molecular integrity. Single-cell morphometrics revealed cardiomyocyte heterogeneity and demonstrated that swimming exercise induces both hypertrophy and hyperplasia, while ventricular injury triggers expansion of regenerative gata4 cardiomyocytes. We integrated proteomics and RNA-seq from FACS-purified cells to construct cell-type-specific proteo-transcriptomic atlases. Functional enrichment, transcription factor, and network analyses identified protein hubs and regulatory circuits defining cardiomyocyte and endothelial cell identity. Our platform delivers cell-type-resolved molecular datasets of adult teleost cardiac cells, establishing a systems-level resource for heart regeneration research, cardiovascular disease modeling, and drug discovery. TeaserA robust workflow for isolation, FACS, imaging, and multi-omics profiling of cardiomyocytes and cardiac endothelial cells in fish. HIGHLIGHTSO_LITemperature-optimized dissociation and standardized FACS enable high-yield isolation of viable cardiac cells. C_LIO_LISingle-cell imaging uncovers morphological heterogeneity in adult ventricular cardiomyocytes. C_LIO_LISustained exercise induces both hypertrophy and hyperplasia of zebrafish cardiomyocytes. C_LIO_LIProteo-transcriptomics defines core molecular programs and interaction networks in cardiomyocytes and endothelial cells. C_LI

systems biology↗

bioRxiv2025

Generative semantic multiplexing (SemaPlex) for accessible and scalable multiplexed fluorescence imaging

Multiplexed fluorescence imaging enhances spatially-resolved interrogation of complex, multi-molecular cell processes that are insufficiently sampled using standard 4-5 plex imaging. To improve accessibility and scalability for multiplexed imaging, we demonstrate generative Semantic Multiplexing (SemaPlex); a simple experimental and deep learning strategy for amplifying marker plexity several-fold by semantically unmixing multiple markers combined per imaging channel. We first characterise key determinants of SemaPlex performance, achieving precise computational multiplexing of 2-to-8 markers synthetically mixed in one channel, facilitating enhanced cell phenotype classification. We then demonstrate practical SemaPlex application, acquiring 10 markers over 4 channels (3*3-plex+1) to efficiently emulate real multiplexed labelling. This permitted accurate reconstruction of quantitative single-cell phenotypic manifolds delineating cell-cycle and mitotic dynamics, with internally validated error-detection. Finally, we exemplify use of semantic guides; additional input channels that significantly enhance multiplexing fidelity. SemaPlex makes multiple-fold increases in fluorescence imaging-plexity accessible, scalable and customisable; democratising multiplexed imaging-based interrogation of complex cell biology.

systems biology↗

bioRxiv2025

Quorum Sensing Regulators and Non-ribosomal Peptide Synthetases Govern Antibacterial Secretions in Xenorhabdus szentirmaii

The decades-long gap in antibiotic discovery has led to a significant health crisis due to antimicrobial resistance (AMR). The bacterial genus Xenorhabdus, which forms symbiotic relationships with the soil nematode Steinernema, are known to secrete a variety of antimicrobial compounds with potential effectiveness against AMR. These antimicrobial compounds are primarily bio-synthesized by non-ribosomal peptide synthetases (NRPS) and polyketide synthetase (PKS) genes. In this study, we report that X. szentirmaii produces high levels of antibiotic activity during the stationary phase against diverse bacteria including known antibiotic resistant pathogens. It possesses 17 operons to encode predicted NRPS and PKS enzymes, designated as ste1 through ste17. The ste15-ste16 and ste17 operons are predicted to produce the known antibiotics Pax peptide and Fabclavine, respectively. Additionally, the newly identified operons ste3, ste4, ste5, ste8, ste9, and ste14 consist of single genes, each containing two or more NRPS genes. The ste13 operon harbors two NRPS genes, while the ste7 and ste12 operons contain three NRPS genes each. Further, RNA-seq analysis showed that lsrF that encodes a quorum sensing autoinducer-2 (AI-2) aldolase was expressed at high levels during stationary phase. These findings provide evidence that X. szentirmaii uses quorum sensing (QS) to synchronize the expression of multiple NRPS and PKS enzymes responsible for synthesizing various antimicrobial compounds. This study underscores the potential to leverage these regulatory insights for maximizing commercial applications of novel antibiotics combating AMR, as well as broader industrial uses.

microbiology↗

bioRxiv2024

CASSIA allows for robust, automated cell annotation in single-cell RNA-sequencing data

Cell type annotation is an essential step in single-cell RNA-sequencing analysis, and numerous annotation methods are available. Most require a combination of computational and domain-specific expertise, and they frequently yield inconsistent results that can be challenging to interpret. Large language models have the potential to expand accessibility while reducing manual input and improving accuracy, but existing approaches suffer from hyperconfidence, hallucinations, and lack of reasoning. To address these limitations, we developed CASSIA for automated, accurate, and interpretable cell annotation of single-cell RNA-sequencing data. As demonstrated in analyses of over 970 cell types, CASSIA improves annotation accuracy in benchmark datasets as well as complex and rare cell populations, and also provides users with reasoning and quality assessment to ensure interpretability, guard against hallucinations, and calibrate confidence.

bioinformatics↗

bioRxiv2023

Adaptation to ER Stress by Slt2, Counterpart of Human MAP Kinase ERK1/2, via Enhancing Splicing and Translation of HAC1 mRNA in Saccharomyces cerevisiae

Unfolded protein response (UPR) is a cellular strategy to increase the protein folding capacity of cells in response to stress within the endoplasmic reticulum (ER). In metazoan cells, three major UPR sensors Ire1, PERK and ATF6 work in concert by simultaneously activating intracellular signaling pathways and modulating a series of physiological processes such as attenuation of the general protein synthesis and expression of protein chaperones. In yeast Saccharomyces cerevisiae, Ire1 is known to be the only UPR sensor, which mediates splicing of HAC1 mRNA in the cytoplasm and derepresses its translation. Hac1 is a transcription factor that increases the expression of protein folding enzymes and chaperones, thus enhancing the protein folding capacity of cells. In this study, we provide compelling evidence that kinase Slt2 plays a significant role in facilitating both the splicing and translation of HAC1 mRNA, while also serving as a key mediator in the activation of UPR genes through an alternative route. We also provide evidence that human extracellular signal-regulated kinase 1 (ERK1) or ERK2 served as a functional substitute for yeast Slt2 in the context of UPR. Furthermore, ERK1 exhibits an enhanced activation in human primary cells when grown in the presence of ER stressor. These findings collectively suggest that Slt2 responds to ER stress by activating the Ire1 pathway as well as initiating a parallel signaling pathway.

molecular biology↗

bioRxiv2022

3D Bioprinted perfusable and vascularized breast tumor model for dynamic screening of chemotherapeutics and CAR-T cells

Despite substantial advancements in development of cancer treatments, lack of standardized and physiologically-relevant in vitro testing platforms limit the rapid and early screening of anti-cancer agents. A major barrier in this endeavor, is the complex interplay between the tumor microenvironment and host immune response and lack of predictive biomarkers for clinical benefit. To tackle this challenge, we have developed a dynamic-flow based three-dimensionally (3D) bioprinted vascularized breast tumor model, responding to chemo and immunotherapeutic treatments. Heterotypic tumor spheroids, comprising metastatic breast cancer cells (MDA-MB-231), human umbilical vein endothelial cells (HUVECs) and human dermal fibroblasts (HDFs), precisely bioprinted at pre-defined distances from a perfused vasculature, exhibited tumor angiogenesis and cancer invasion. Proximally bioprinted tumors ([~]100 m) exhibited enhanced capillary sprouting, anastomosis to perfused vasculature and increased cancer cell migration as compared to distally bioprinted spheroids ([~]500 m). Proximally bioprinted tumors treated with varying dosages of doxorubicin for 72 h enabled functional analysis of drug response, wherein, tumors portrayed a dose-dependent drug response behavior with [~]70% decrease in tumor volume for 1 M dose. Additionally, a cell based immune therapy approach was explored by perfusing HER2-targeting chimeric antigen receptor (CAR) modified CD8+ T cells for 24 or 72 h through the central vasculature. Extensive CAR-T cell recruitment to the endothelium and substantial T cell activation and infiltration in the tumor site, resulted in [~]70% reduction in tumor growth for high CAR treatment densities, after 72 h of treatment. The presented 3D model paves the way for a robust, precisely fabricated and physiologically-relevant 3D tumor microenvironment platform for future translation of anti-cancer therapies to personalized medicine for cancer patients. One Sentence SummaryA physiologically-relevant 3D bioprinted perfusable vascularized tumor model capable of dynamic screening of chemo and immunotherapeutics.

bioengineering↗

bioRxiv2022

Development of 3D breast cancer models with human T cells expressing engineered MAIT cell receptors

Immunotherapy has revolutionized cancer treatment with the advent of advanced cell engineering techniques aimed at targeted therapy with reduced systemic toxicity. However, understanding the underlying immune-cancer interactions require development of advanced three-dimensional (3D) models of human tissues. In this study, we developed proof-of-concept 3D tumor models with increasing complexity to study the cytotoxic responses of CD8+ T cells, genetically engineered to express mucosal-associated invariant T (MAIT) cell receptors, towards MDA-MB-231 breast cancer cells. Homotypic MDA-MB-231 and heterotypic MDA-MB-231/human dermal fibroblast (HDF) tumor spheroids were primed with precursor MAIT cell ligand 5-amino-6-D-ribitylaminouracil (5-ARU). Engineered T cells effectively eliminated tumors after a 3-day culture period, demonstrating that the engineered TCR recognized MR1 expressing tumor cells in the presence of 5-ARU. Tumor cell killing efficiency of engineered T cells were also assessed by encapsulating these cells in fibrin, mimicking a tumor extracellular matrix microenvironment. Expression of proinflammatory cytokines such as IFN{gamma}, IL-13, CCL-3 indicated immune cell activation in all tumor models, post immunotherapy. Further, in corroborating the cytotoxic activity, we found that granzymes A and B were also upregulated, in homotypic as well as heterotypic tumors. Finally, a 3D bioprinted tumor model was employed to study the effect of localization of T cells with respect to tumors. T cells bioprinted proximal to the tumor had reduced invasion index and increased cytokine secretion, which indicated a paracrine mode of immune-cancer interaction. Development of 3D tumor-T cell platforms may enable studying the complex immune-cancer interactions and engineering MAIT cells for cell-based cancer immunotherapies.

immunology↗

bioRxiv2020

RNA-binding protein FXR1 drives cMYC translation by mRNA circularization through eIF4F recruitment in ovarian cancer

BackgroundThe RNA-binding protein FXR1 (fragile X-related protein 1) has been implicated as an important regulator of post-transcriptional changes of mRNAs. However, its role in mRNA circularization and recruitment of eukaryotic translation initiation factors for protein translation remains obscure. Here, we aimed to investigate the molecular mechanisms and potential clinical applications of FXR1 in ovarian cancer growth and progression. MethodsFXR1 copy number variation, mRNA expression, protein levels, and their association with prognosis were determined in clinical datasets. An orthotopic ovarian cancer model and bioluminescence imaging were used for preclinical evaluation of FXR1 in vivo. Reverse phase protein arrays (RPPA) and qPCR arrays were performed to identify FXR1s key targets and downstream effects. SUnSET and polysome profiling were used to determine the translational effects of FXR1. Immunoprecipitation and immunofluorescence were performed to identify the interaction between FXR1 and cMYC mRNA and eIF4F complex. RNA-immunoprecipitation (RIP), RNA electrophoretic mobility shift assays (REMSA), proximity ligation assays (PLA), and biochemical assays were used to identify the specific site on cMYC mRNA to which FXR1 binds to promote mRNA circularization and translation. ResultsWe found that amplification and copy-gain of FXR1 increased the expression of FXR1 mRNA and FXR1 protein in ovarian cancer patients, and these events associated with poor prognosis. We demonstrated that FXR1 binds to AU-rich elements (ARE) within the 3 untranslated region (3UTR) of cMYC. As a consequence, FXR1 binding to cMYC 3UTR leads to the circularization of mRNA and facilitated the recruitment of eukaryotic translation initiation factors (eIFs) to translation start site for improving protein synthesis. ConclusionWe found that FXR1 upregulates a known oncogene, cMYC, by binding to AU-rich elements within the 3UTR, leading to the recruitment of the eIF4F complex for cMYC translation. Our findings uncover a novel mechanism of action of FXR1 in tumorigenesis and provides opportunities to use FXR1 and its downstream effectors as biomarkers or therapeutic targets in ovarian and other cancers.

cancer biology↗

bioRxiv2020

3D bioprinting of co-cultured osteogenic spheroids for bone tissue fabrication

Conventional top-down approaches in tissue engineering involving cell seeding on scaffolds have been widely used in bone engineering applications. However, scaffold-based bone tissue constructs have had limited clinical translation due to constrains in supporting scaffolds, minimal flexibility in tuning scaffold degradation, and low achievable cell seeding density as compared with native bone tissue. Here, we demonstrate a pragmatic and scalable bottom-up method, inspired from embryonic developmental biology, to build three-dimensional (3D) scaffold-free constructs using spheroids as building blocks. Human umbilical vein endothelial cells (HUVECs) were introduced to human mesenchymal stem cells (hMSCs) (hMSC/HUVEC) and spheroids were fabricated by an aggregate culture system. Bone tissue was generated by induction of osteogenic differentiation in hMSC/HUVEC spheroids for 10 days, with enhanced osteogenic differentiation and cell viability in the core of the spheroids compared to hMSC-only spheroids. Aspiration-assisted bioprinting (AAB) is a new bioprinting technique which allows precise positioning of spheroids (11% with respect to the spheroid diameter) by employing aspiration to lift individual spheroids and bioprint them onto a hydrogel. AAB facilitated bioprinting of scaffold-free bone tissue constructs using the pre-differentiated hMSC/HUVEC spheroids. These constructs demonstrated negligible changes in their shape for two days after bioprinting owing to the reduced proliferative potential of differentiated stem cells. Bioprinted bone tissues showed interconnectivity with actin-filament formation and high expression of osteogenic and endothelial-specific gene factors. This study thus presents a viable approach for 3D bioprinting of complex-shaped geometries using spheroids as building blocks, which can be used for various applications including but not limited to, tissue engineering, organ-on-a-chip and microfluidic devices, drug screening and, disease modeling.

bioengineering↗

bioRxiv2020

Ir56d-dependent fatty acid responses in Drosophila uncovers taste discrimination between different classes of fatty acids

Chemosensory systems are critical for evaluating the caloric value and potential toxicity of food prior to ingestion. While animals can discriminate between 1000s of odors, much less is known about the discriminative capabilities of taste systems. Fats and sugars represent calorically potent and innately attractive food sources that contribute to hedonic feeding. Despite the differences in nutritional value between fats and sugars, the ability of the taste system to discriminate between different rewarding tastants is thought to be limited. In Drosophila, sweet taste neurons expressing the Ionotropic Receptor 56d (IR56d) are required for reflexive behavioral responses to the medium-chain fatty acid, hexanoic acid. Further, we have found that flies can discriminate between a fatty acid and a sugar in aversive memory assays, establishing a foundation to investigate the capacity of the Drosophila gustatory system to differentiate between various appetitive tastants. Here, we tested whether flies can discriminate between different classes of fatty acids using an aversive memory assay. Our results indicate that flies are able to discriminate medium-chain fatty acids from both short- and long-chain fatty acids, but not from other medium-chain fatty acids. Characterization of hexanoic acid-sensitive Ionotropic receptor 56d (Ir56d) neurons reveals broad responsive to short-, medium-, and long-chain fatty acids, suggesting selectivity is unlikely to occur through activation of distinct sensory neuron populations. However, genetic deletion of IR56d selectively disrupts response to medium chain fatty acids, but not short and long chain fatty acids. These findings reveal Ir56d is selectively required for fatty acid taste, and discrimination of fatty acids occurs through differential receptor activation within shared populations of neurons. These findings uncover a capacity for the taste system to encode tastant identity within a taste category.

neuroscience↗

bioRxiv2018

Interleukin-8/CXCR2 signaling regulates therapy-induced plasticity and enhances tumorigenicity in glioblastoma

Glioblastoma (GBM) remains one of the least treatable types of cancer. Recent work highlights two key factors contributing to this resistant phenotype--cellular plasticity, the ability of GBM cells to adopt many phenotypes, and the microenvironment. Here, we provide evidence that Interleukin-8 (IL-8) plays a vital role in promoting cellular plasticity and cancer initiating cells (CICs) niche during anti-glioma chemotherapy. IL-8 expression is significantly elevated during chemotherapy, and immunohistochemical analysis of matched primary and recurrent patient GBM tissues revealed about 60% of recurrent tissues IL-8 expression is upregulated. In silico analysis of the TCGA data indicated that IL-8 signaling could promote epigenetic plasticity by altering the polycomb repressor complex activity. We are proposing that such regulation my promote epigenetic plasticity, which allows the GBM cells to adapt therapy and may promote therapeutic resistance. Our data show that IL-8 is a crucial microenvironmental factor involved in developing therapeutic adaptation and can be targeted in combination with conventional chemo-and radiotherapy to prevent disease recurrence.\n\nABSTRACTEmerging evidence reveals enrichment of glioma initiating cells (GICs) following therapeutic intervention. This enrichment occurs partly by dedifferentiation of non-GICs to GICs within the tumor, which may contribute to therapeutic resistance and the generation of lethal recurrent tumors. To elucidate the molecular mechanisms governing therapy-induced cellular plasticity, we performed genome-wide chromatin immunoprecipitation sequencing (ChIP-Seq) and gene expression analysis (gene microarray analysis) during treatment with standard of care temozolomide (TMZ) chemotherapy. Analysis revealed significant enhancement of open chromatin marks in known astrocytic enhancers for Interleukin-8 (IL-8) loci as well as elevated expression during anti-glioma chemotherapy. The Cancer Genome Atlas and Ivy Glioblastoma Atlas Project data demonstrated that IL-8 transcript expression is negatively correlated with GBM patient survival (p=0.001) and positively correlated with that of genes associated with the CIC phenotype such as KLF4, c-Myc and HIF2 (p<0.001). Immunohistochemical analysis of patient samples demonstrated elevated IL-8 expression in about 60% of recurrent GBM tumors relative to matched primary tumors and this expression also positively correlates with time to recurrence. Exposure to IL-8 significantly enhanced the self-renewing capacity of patient-derived xenograft (PDX) GBM (average 3-fold, p<0.0005). Furthermore, IL-8 knockdown significantly delayed PDX GBM tumor growth in vivo (p<0.0005). Finally, guided by in silico analysis of TCGA data, we examined the effect of therapy-induced IL-8 expression on the epigenomic landscape of GBM cells and observed increased trimethylation of H3K9 and H3K27. Our results show that IL-8 alters cellular plasticity and mediates alterations in histone status. These finding suggest that IL-8 signaling participates in regulating GBM adaptation to therapeutic stress and therefore represents a promising target for combination with conventional chemotherapy in order to limit GBM recurrence.

cancer biology↗