Biology subjects

Roy, G.

Publications and source records attributed to Roy, G..

11 recordsLinked to original sources

bioRxiv2026

A Distinct Mode of Chromosome-End Genome Instability Revealed by Circulome Characterization and Analyses

Genome instability is a central hallmark of cancer and it is characterized largely from sequence: mutation, copy-number change and rearrangement scored against a linear reference. The genome has additional degrees of freedom, such as the physical form (topology) of the DNA itself. A specific mode of localized genome dynamics changes both dimensions (sequence and topology) at once: defined chromosomal regions are also represented as extrachromosomal circular DNA (eccDNA). The biogenesis of eccDNA is a feature of many cancers, supplying the tumor with genetic elements that can participate in phenotypic selection, and supplying the observer with a window into the genomic instability landscape. Genome-wide maps of where these circles arise have not been available at scale or at high resolution. Building such maps requires physical selection of circular species, because most sequenced reads from unselected DNA derive from unrearranged sequence, conflating the signal from dynamic rearrangements with that from bulk chromosomal DNA. Here we combine biophysical and biochemical isolation of circular DNA with next-generation sequencing, applying it to a panel of cancer cell lines and to patient tumors, with non-cancerous cells and adjacent normal tissues as controls. The resulting genome-wide maps of the circular-element repertoire, the Circulome, reveal a consistent and previously unreported enrichment of these elements at particular chromosome ends. The circles map to regions extending well past the telomeres, reaching varying distances up to tens of megabases inward. This activity is accompanied by corroborating rearrangements within the linear genome, consistent with regions of chromosomal instability. Together these features define a spatially localized mode of instability in cancer, which we term Erosion at Distal Genomic Ends (EDGE). We formulate a classification problem to discriminate bladder tumor from matched normal tissue, finding that the EDGE signal outperforms the mean of all interior chromosomal regions, for each chromosome. These findings define a distinct spatial signature of cancer chromosome instability detectable through Circulome analysis, and connect early models of region-specific instability to the human cancer genome.

genomics↗

bioRxiv2026

Human CD4 T cells coordinate antitumor immunity through indirect antigen presentation

Whether human tumor-specific CD4 T cells can drive antitumor immunity in cancers lacking tumor-cell MHC class II expression has not been established. In MHC class II-negative melanoma and breast cancer, CD4 T cells with a transcriptional signature of tumor specificity correlated with myeloid and CD8 T cell activation. Using a patient-derived NY-ESO-specific, HLA-DP04-restricted TCR, we tested whether human CD4 T cells could control tumors through indirect antigen presentation. In vitro, CD4 T cells inhibited HLA-DP-negative melanoma growth through recognition of tumor-derived antigen on antigen presenting cells (APC). In humanized mice, CD4 T cell transfer induced regression of HLA-DP-negative tumors, and control required HLA-matched human APCs. CD4 T cell transfer induced endogenous CD8 T cell proliferation and myeloid activation that mirrored immune programs correlated with the presence of these cells in patients. These findings establish that human tumor-specific CD4 T cells coordinate antitumor immunity through APC without direct tumor recognition, supporting their therapeutic relevance in MHC class II-negative cancers.

immunology↗

bioRxiv2025

Unique and conserved endoplasmic reticulum stress responses in neuroendocrine cells

Endocrine cells are dedicated to the production and processing of hormones, from peptides to small molecules, to regulate key physiological processes, including glucose homeostasis and metabolism. Because of this relatively high productivity, endo-crine cells must handle a variety of stresses from oxidative stress to the unfolded protein response of the endoplasmic reticulum (UPRER). While much is known about the major pathways regulating the UPRER, the roles of endocrine cell type-specific, context-dependent, and time-dependent transcriptional changes are not well explored. To identify unique and shared responses to the UPRER across a subset of endocrine cell types, we tested representative lines for {beta}-cells (insulin), -cells (glucagon), {delta}-cells (somatostatin), X/A-cells (ghrelin), L-cells (glucagon-like peptide 1 (GLP1)), and thyrotropes (thyroid hormone and thyroglobulin). We exposed each cell type to the canonical ER stressor thapsigargin for 6 and 24 h, or vehicle for 24 h and performed mRNA sequencing. Analysis of the data showed all lines responded to thapsigargin. Comparisons of differentially expressed genes between each line revealed both shared and unique transcriptional signatures. These data represent a valuable mineable set of candidate genes that may have cell type-specific functions during the UPRER and have the potential to lead to a new understanding of how different endocrine cells mitigate or succumb to ER stress.

cell biology↗

bioRxiv2025

MetagenBERT: a Transformer Architecture using Foundational DNA Read Embedding Models to enhance Disease Classification

Microbial ecosystems constitute complex yet information-rich environments whose characterization is crucial for understanding host health and disease. Among them, the human gut microbiome has emerged as a key "super-integrator", owing to its dense interactions with host physiology and its established associations with a wide spectrum of pathologies. Driven by advances in high-throughput sequencing technologies and the continuous decline in associated costs, metagenomic studies have expanded exponentially, generating massive amounts of sequencing data and opening new avenues for data-driven disease modeling. Conventional approaches to microbiome analysis predominantly rely on the alignment of DNA sequencing reads against reference databases to infer microbial composition and profiling at the species level. While effective, these methods are inherently constrained by reference bias and limited taxonomic resolution. Recent advances in artificial intelligence--particularly in Natural Language Processing (NLP) offer new methodological perspectives for metagenomic data representation. In this study, we present MetagenBERT, a Transformer-based framework to embed metagenomes that relies on the foundational models DNABERT-2 and DNABERT-S for the embedding of DNA sequencing reads. Our approach encodes gut microbiome metagenome in a taxonomy-agnostic manner, enabling direct downstream application to disease classification tasks. We demonstrate that MetagenBERT reaches similar performance to state-of-the-art abundance-based models for cirrhosis prediction and surpasses them in the more challenging context of type 2 diabetes. Furthermore, we introduce an alternative representation of metagenomes based on read-level embeddings aggregated into abundance vectors, demonstrating their complementarity with conventional species-level abundance metrics.

bioinformatics↗

bioRxiv2024

Target deconvolution of an insulin hypersecretion-inducer acting through VDAC1 with a distinct transcriptomic signature in beta-cells

Obesity, insulin resistance, and a host of environmental and genetic factors can drive hyperglycemia, causing {beta}-cells to compensate by increasing insulin production and secretion. In type 2 diabetes (T2D), {beta}-cells under these conditions eventually fail. Rare {beta}-cell diseases like congenital hyperinsulinism (HI) also cause inappropriate insulin secretion, and some HI patients develop diabetes. However, the mechanisms of insulin hypersecretion and how it causes {beta}-cell dysfunction are not fully understood. We previously discovered small molecules (e.g. SW016789) that cause insulin hypersecretion and lead to a loss in {beta}-cell function without cell death. Here, we uncover the protein target of SW016789 and provide the first time-course transcriptomic analysis of hypersecretory responses versus thapsigargin-mediated ER stress in {beta}-cells. In mouse MIN6 and human EndoC-{beta}H1 {beta}-cells, we identified and validated VDAC1 as a SW016789 target using photoaffinity proteomics, cellular thermal shift assays, siRNA, and small molecule inhibitors. SW016789 raises membrane potential to enhance Ca2+ influx, potentially through VDAC1. Chronically elevated intracellular Ca2+ appears to underpin the negative impacts of hypersecretion, as nifedipine protected against each small molecule hypersecretion inducer we tested. Using time- course RNAseq, we discovered that hypersecretion induced a distinct transcriptional pattern compared to ER stress. Clustering analyses led us to focus on ER-associated degradation (ERAD) as a potential mediator of the adaptive response. SW016789 reduced the abundance of ERAD substrate OS-9 and pharmacological inhibition of ERAD worsened {beta}-cell survival in response to hypersecretory stress. Changes in other ERAD components in MIN6 and EndoC-{beta}H1 at the protein level were minor with either SW016789 or thapsigargin. However, immunostaining for core ERAD components SEL1L, HRD1, and DERL3 in non-diabetic and T2D human pancreas revealed altered distributions of SEL1L/HRD1 and SEL1L/DERL3 rations in {beta}-cells of T2D islets, in alignment with altered ERAD in stressed {beta}-cells. We conclude that hypersecretory stimuli, including SW016789- mediated VDAC1 activation, cause enhanced Ca2+ influx and insulin release. Subsequent differential gene expression represents a {beta}-cell hypersecretory response signature that is reflected at the protein level for some, but not all genes. A better understanding of how {beta}-cells induce hypersecretion and the mechanisms of negative feedback on secretory rate may lead to the discovery of novel therapeutic targets for T2D and HI.

cell biology↗

bioRxiv2024

IB-DNQ and Rucaparib dual treatment alters cell cycle regulation and DNA repair in triple negative breast cancer cells

Triple negative breast cancer (TNBC) is a highly aggressive breast cancer that is unresponsive to hormonal therapies. One potential TNBC-specific therapeutic target is NQO1, as it is highly expressed in many TNBC patients and lowly expressed in non-cancer tissues. Here we use a derivative of DNQ, isobutyl-deoxynyboquinone (IB-DNQ) that is more potent and specific in killing TNBC cells than NQO1-activator {beta}-lapachone while displaying strong NQO1-dependence. We evaluated the cellular signaling changes that occur following 4-hour treatment of TNBC cells with either single agent or combination IB-DNQ and / or PARP inhibitor (Rucaparib). Short treatments (4 hours) with IB-DNQ alone or combined with the PARP inhibitor Rucaparib revealed few changes in protein abundance but significant rapid alterations in protein phosphorylation and thermal stability, with clear synergy in the combination treatment. Key phosphorylated targets linked to RNA Polymerase II inhibition and DNA damage response were altered during our short time treatment. Thermal proteome profiling (TPP) identified novel, combination-specific changes in protein biophysical state suggesting new therapeutic vulnerabilities in TNBC cells. Our findings highlight how even brief treatments can uncover distinct biophysical protein changes via TPP, offering a resource for mechanistic studies of IB-DNQ mechanism of action and the development of NQO1-activated therapeutics for TNBC treatment.

cancer biology↗

bioRxiv2024

Identification of type 2 diabetes- and obesity-associated human β-cells using deep transfer learning

Diabetes affects >10% of adults worldwide and is caused by impaired production or response to insulin, resulting in chronic hyperglycemia. Pancreatic islet {beta}-cells are the sole source of endogenous insulin and our understanding of {beta}-cell dysfunction and death in type 2 diabetes (T2D) is incomplete. Single-cell RNA-seq data supports heterogeneity as an important factor in {beta}-cell function and survival. However, it is difficult to identify which {beta}-cell phenotypes are critical for T2D etiology and progression. Our goal was to prioritize specific disease-related {beta}-cell subpopulations to better understand T2D pathogenesis and identify relevant genes for targeted therapeutics. To address this, we applied a deep transfer learning tool, DEGAS, which maps disease associations onto single-cell RNA-seq data from bulk expression data. Independent runs of DEGAS using T2D or obesity status identified distinct {beta}-cell subpopulations. A singular cluster of T2D-associated {beta}-cells was identified; however, {beta}-cells with high obese-DEGAS scores contained two subpopulations derived largely from either non-diabetic or T2D donors. The obesity-associated non-diabetic cells were enriched for translation and unfolded protein response genes compared to T2D cells. We selected CDKN1C and DLK1 for validation by immunostaining in human pancreas sections from healthy and T2D donors. Both CDKN1C and DLK1 were heterogeneously expressed among {beta}-cells. CDKN1C was increased in {beta}-cells from T2D donors, in agreement with the DEGAS predictions, while DLK1 appeared depleted from T2D islets of some donors. In conclusion, DEGAS has the potential to advance our holistic understanding of the {beta}-cell transcriptomic phenotypes, including features that distinguish {beta}-cells in obese non-diabetic or lean T2D states. Future work will expand this approach to additional human islet omics datasets to reveal the complex multicellular interactions driving T2D.

bioinformatics↗

bioRxiv2023

Deep Learning Methods in Metagenomics: a systematic review

The ever-decreasing cost of sequencing and the growing potential applications of metagenomics have led to an unprecedented surge in data generation. One of the most prevalent applications of metagenomics is the study of microbial environments, such as the human gut. The gut microbiome plays a crucial role in human health, providing vital information for patient diagnosis and prognosis. However, analyzing metagenomic data remains challenging due to several factors, including reference catalogs, sparsity, and compositionality. Deep learning (DL) enables novel and promising approaches that complement state-of-the-art microbiome pipelines. DL-based methods can address almost all aspects of microbiome analysis, including novel pathogen detection, sequence classification, patient stratification, and disease prediction. Beyond generating predictive models, a key aspect of these methods is also their interpretability. This article reviews deep learning approaches in metagenomics, including convolutional networks (CNNs), autoencoders, and attention-based models. These methods aggregate contextualized data and pave the way for improved patient care and a better understanding of the microbiomes key role in our health. Author summaryIn our study, we look at the vast world of research in metagenomics, the study of genetic material from environmental samples, spurred by the increasing affordability of sequencing technologies. Our particular focus is the human gut microbiome, an environment teeming with microscopic life forms that plays a central role in our health and well-being. However, navigating through the vast amounts of data generated is not an easy task. Traditional methods hit roadblocks due to the unique nature of metagenomic data. Thats where deep learning (DL), a today well known branch of artificial intelligence, comes in. DL-based techniques complement existing methods and open up new avenues in microbiome research. Theyre capable of tackling a wide range of tasks, from identifying unknown pathogens to predicting disease based on a patients unique microbiome. In our article, we provide a very comprehensive review of different DL strategies for metagenomics, including convolutional networks, autoencoders, and attention-based models. We are convinced that these techniques significantly enhance the field of metagenomic analysis in its entirety, paving the way for more accurate data analysis and, ultimately, better patient care. The PRISMA augmented diagram of our review is illustrated in Fig 1. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=160 SRC="FIGDIR/small/552187v2_fig1.gif" ALT="Figure 1"> View larger version (18K): org.highwire.dtl.DTLVardef@9f0060org.highwire.dtl.DTLVardef@d1cd1aorg.highwire.dtl.DTLVardef@1f0a5fborg.highwire.dtl.DTLVardef@211b5a_HPS_FORMAT_FIGEXP M_FIG O_FLOATNOFig 1.C_FLOATNO PRISMA-type diagram for article selection of this review. The method developed here enriches the research equation selection with Connected Papers, this diagram represents the selection along with this enrichment in green C_FIG

bioinformatics↗

bioRxiv2022

Post-transcriptional regulation by copper with a new upstream Open Reading Frame

Copper is essential to most living beings but also toxic. Bacteria have thus developed homeostatic mechanisms to tightly control its intracellular concentration. The 3-gene operon bp2923-bfrG-bp2921 is down-regulated by copper and notably encodes a TonB-dependent transporter in Bordetella pertussis. We show that the protein encoded by bp2923, which is a member of the DUF2946 family, represents a new type of upstream Open Reading Frame (uORF) involved in post-transcriptional regulation of the downstream genes. In the absence of copper, the entire operon is transcribed and translated. Perception of copper by the nascent bp2923-coded protein via its conserved CXXC motif triggers Rho-dependent transcription termination between the first and second genes by relieving translation arrest on a conserved C-terminal RAPP motif. Homologues of bp2923 are widespread in bacterial genomes, where they head operons predicted to participate in copper homeostasis. This work has unveiled an original mode of genetic regulation by a transition metal and identified a regulatory function for a member of an uncharacterized family of bacterial proteins that we have named CruR, for copper-responsive upstream regulator.

microbiology↗

bioRxiv2022

Small molecule-mediated insulin hypersecretion induces transient unfolded protein response and loss of function in pancreatic islet beta cells

Pancreatic islet beta cells require a fine-tuned ER stress response for normal function; abnormal ER stress contributes to diabetes pathogenesis. Here, we identified a small molecule, SW016789, with time-dependent effects on beta cell ER stress and function. Acute treatment with SW016789 potentiated nutrient-induced calcium influx and insulin secretion, while chronic exposure to SW016789 transiently induced ER stress and shut down secretory function in a reversible manner. Distinct from the effects of thapsigargin, SW016789 did not affect beta cell viability or apoptosis, potentially due to a rapid induction of adaptive genes, weak signaling through the eIF2 kinase PERK, and lack of oxidative stress gene Txnip induction. We determined that SW016789 acted upstream of voltage-dependent calcium channels (VDCCs) and potentiated nutrient- but not KCl-stimulated calcium influx. Measurements of metabolomics, oxygen consumption rate, and G protein-coupled receptor signaling did not explain the potentiating effects of SW016789. In chemical co-treatment experiments we discovered synergy between SW016789 and activators of protein kinase C (PKC) and VDCCs, suggesting involvement of these pathways in the mechanism of action. Finally, chronically elevated calcium influx was required for the inhibitory impact of SW016789, as blockade of VDCCs protected human islets and MIN6 beta cells from hypersecretion-induced dysfunction. We conclude that beta cells undergoing this type of pharmacological hypersecretion have the capacity to suppress their function to mitigate ER stress and avoid apoptosis. These results have the potential to uncover beta cell ER stress mitigation factors and add support to beta cell rest strategies to preserve function.

cell biology↗

bioRxiv2021

Warming in the land of the midnight sun: breeding birds may suffer greater heat stress at high- vs low-Arctic sites

Rising global temperatures are expected to increase reproductive costs for wildlife as greater thermoregulatory demands interfere with essential breeding activities such as parental care. However, predicting the temperature threshold where reproductive performance is negatively impacted remains a significant hurdle. Using a novel thermoregulatory polygon approach, we predicted the threshold temperature at which an Arctic songbird-the snow bunting (Plectrophenax nivalis)-would need to reduce activity and perform below the 4-times basal metabolic rate (BMR) required to sustain nestling provisioning to avoid overheating. We then compared this threshold to operative temperatures recorded at high (82{degrees}N) and low (64{degrees}N) Arctic sites to estimate how heat constraints translate into site-specific impacts on sustained activity level. We predict buntings would become behaviourally constrained at operative temperatures above 11.7{degrees}C, whereupon they must reduce provisioning rates to maintain thermal balance. Low Arctic sites had larger fluctuations in solar radiation, producing consistent daily periods when operative temperatures exceeded 11.7{degrees}C. However, high-latitude birds faced entire, consecutive days where parents would not be able to sustain required provisioning rates. These data indicate that Arctic warming is likely already disrupting the breeding performance of cold-specialist birds, but also suggests counterintuitive and severe negative impacts of warming at high-latitude breeding locations.

ecology↗