Biology subjects

Bhattacharya, S.

Publications and source records attributed to Bhattacharya, S..

At least 37 records · Page 2Linked to original sources

bioRxiv2026

Mapping the Ligand-dependent Remodeling of the Conformational Entropy Landscape in Neurotensin Receptor 1 by NMR-guided Molecular Simulations

This study presents a comprehensive analysis of the dynamic properties and allosteric regulation mechanisms of Class A G protein-coupled receptors (GPCRs) by integrating molecular dynamics (MD) simulations with nuclear magnetic resonance (NMR) relaxation measurements. Utilizing generalized order parameters derived from NMR data and MD trajectories, we quantitatively assess conformational entropy changes that occur during receptor activation and ligand binding events. This approach enables a detailed characterization of protein flexibility at multiple timescales, revealing how dynamic fluctuations contribute to allosteric signal transmission within the receptor. Our results demonstrate that conformational entropy plays a pivotal role in modulating the functional states of Class A GPCRs, influencing the equilibrium between inactive and active conformations. By elucidating the interplay between structural dynamics and allostery, this work advances the molecular-level understanding of GPCR function and highlights the importance of entropy-driven effects in receptor signaling. The integrative methodology and findings provide a valuable framework for future investigations aimed at targeting receptor dynamics in drug discovery and rational design of allosteric modulators.

molecular biology↗

bioRxiv2025

Unveiling Differential Epigenetic Regulation of ABC Transporter Genes in Wild and Domestic Rice in Response to Heavy Metal Stress

Plants are continually subjected to heavy metal (HM) stress, which impairs critical physiological and metabolic processes. ATP-binding cassette (ABC) transporters play an important role in heavy metal detoxification, their epigenetic regulation under heavy metal stress remains poorly understood. Epigenetic changes, such as DNA methylation (5mC) and N6-methyladenine (6mA), are emerging as important regulators of stress-responsive genes. We studied the epigenetic regulation of ABC transporter genes in three Oryza species Oryza sativa indica, Oryza rufipogon, and Oryza glaberrima by analysing promoter methylation patterns. Our data show that 5mC is abundant in CHG motifs, whereas 6mA is mostly found at AHCA motifs, indicating sequence-specific methylation preferences across different rice species. Notably, indica rice exhibited higher promoter methylation levels compared to O. rufipogon and O. glaberrima, suggesting a potential link between increased DNA methylation and reduced ABC transporter expression, which may contribute to heightened HM sensitivity in cultivated rice. Phylogenetic analysis further highlighted species-specific variations in methylation patterns, supporting the role of epigenetics in shaping HM stress adaptation. These results provide new insights into the regulatory mechanisms governing heavy metal tolerance in rice and open avenues for epigenetic-based crop improvement strategies.

plant biology↗

bioRxiv2025

Developmental control of DNA damage responses in α- and β-cells shapes the selective beta-cell susceptibility in diabetes

Accumulation of DNA damage drives {beta}-cell dysfunction, senescence, and death in Type 1 and Type 2 diabetes. While -cell dysfunction also contributes to disease pathology, they remain remarkably resistant to senescence and cell-death. The mechanisms underlying these differential responses to diabetogenic stress, particularly differences in their DNA damage vulnerability, remain unclear. We demonstrate that replication introduces a window of genomic vulnerability in both - and {beta}-cells during neonatal growth, with -cells exhibiting higher replication rates and DNA damage. We show that neonatal {beta}-cells resolve DNA damage more efficiently during mitosis and favor error-free repair, while -cells compensate for their higher DNA damage vulnerability through increased cellular turnover. Using mouse models of overnutrition and diabetes, we show that {beta}-cells exhibit greater vulnerability to terminal DNA damage and impaired repair capacity under diabetogenic stress, with compensatory replication amplifying this vulnerability. We demonstrate that developmental epigenetic programs shape the differential DNA damage vulnerability of postnatal {beta}- and -cells. Loss of de novo DNA methyltransferase Dnmt3a in pancreatic progenitors selectively increases the DNA damage vulnerability of {beta}-cells from neonatal growth through adulthood. Our findings uncover novel developmental mechanisms that shape the distinct DNA damage responses of postnatal {beta}- and -cells during growth and diabetes. Article HighlightsO_LIMechanisms underlying the differential susceptibility of pancreatic {beta}- and -cells to diabetogenic stress remain unclear. C_LIO_LIDo replication dynamics, repair fidelity, and developmental epigenetic programs determine the vulnerability of postnatal {beta}- and -cells to DNA damage, a key driver of {beta}-cell failure in diabetes? C_LIO_LIReplication introduces DNA damage vulnerability in both neonatal {beta}- and -cells, yet {beta}-cells resolve damage more efficiently. Loss of DNA methyltransferase 3a in pancreatic progenitors selectively heightens {beta}-cell DNA damage vulnerability that persists into adulthood. Moreover, diabetogenic-stress preferentially compromises {beta}-cell repair fidelity. C_LIO_LIThese findings reveal how developmental programs shape {beta}-cell resilience and may influence lifelong diabetes risk. C_LI

developmental biology↗

bioRxiv2025

Synergy unleashed: Leishmania donovani GP63 paralogues cooperatively function to drive parasite infectivity at various pathogenic stages and regulate host survival mechanisms

Highly plastic genome of Leishmania is thought to underlie its remarkable ability to adapt to diverse host microenvironments, yet how this plasticity translates into dermal versus visceral persistence remains unclear. GP63, a zinc metalloprotease, is a well-established determinant of promastigote infectivity, particularly in cutaneous leishmaniasis. Despite this, its role in amastigote stage and in visceral disease caused by Leishmania donovani remains undefined. Here, using an integrated approach combining phylogenetics, CRISPR-mediated genome editing, and biochemical analyses, we demonstrate that GP63 paralogues in L. donovani have undergone independent evolutionary diversification to support visceral infection. While GP63-paralogues encoded on chromosome 10, previously implicated in cutaneous manifestation caused by L. major, were found severely truncated and dispensable in L. donovani, two distinct GP63-paralogues encoded on chromosome 28 and 31 were found essential to establish visceral infection. LdGP63_31 facilitates promastigote attachment to macrophages by promoting lipid raft engagement and complement inactivation, thereby enabling host entry and subsequent amastigote differentiation. LdGP63_28 is essential for intracellular amastigote survival by suppressing host inflammatory pyroptosis. Structural and enzymatic analyses revealed preferential host localization and substrate specificities, possibly resulting from distinct amino acid substitution within conserved motifs of these proteases, indicating independent evolutionary adaptations. Importantly, loss of the LdGP63_28 more severely impaired amastigote reinfection than loss of its counterpart LdGP63_31. Together, these findings reveal coordinated functional specialization of GP63-paralogues ensuring visceral adaptation.

microbiology↗

bioRxiv2025

Spatiotemporal regulation of arbuscular mycorrhizal symbiosis at cellular resolution

Arbuscular mycorrhizal (AM) symbiosis develops through successive colonization of root epidermal and cortical cells, culminating in the formation of arbuscules, tree-like intracellular structures that are transient yet essential sites of nutrient exchange. To dissect the cellular and structural complexity of AM establishment in rice roots colonized by Rhizophagus irregularis, we applied dual-species spatial transcriptomics to simultaneously monitor plant and fungal gene transcripts at single-cell resolution. This approach revealed surprising differences in transcriptional activity between fungal structures and showed that morphologically similar arbuscules can be transcriptionally distinct. These findings suggest hidden functional diversity among arbuscules at single-cell resolution. Because arbuscules form and degenerate within only a few days, we further sought to capture translational activities across their life span. We pioneered AM-inducible TRAP-seq (Translating Ribosome Affinity Purification followed by RNA-seq) using stage-specific promoters, enabling cell-type- and stage-resolved profiling in AM symbiosis. This revealed extensive spatiotemporal reprogramming of nutrient transport and signalling, with distinct sets of phosphate, nitrogen, and carbon transporters and regulators induced or repressed at different stages of arbuscule development, suggesting that nutrient exchange is dynamically regulated across the arbuscule life cycle. More broadly, cell wall biosynthesis genes and key defence markers were suppressed during arbuscule formation, whereas at a later stage, defence markers were strongly upregulated, suggesting a host-driven shift towards arbuscule termination. Together, these findings highlight the nuanced and dynamic regulation of AM symbiosis at the cellular level, refining our understanding of how nutrient exchange and fungal development are coordinated in space and time.

plant biology↗

bioRxiv2025

A family of endonucleases that block nanotube-mediated plasmid dissemination

Small non-conjugative plasmids constitute a substantial portion of the bacterial mobile genome, driving the dissemination of beneficial genes, with their transfer primarily attributed to transformation, transduction, or co-mobilization with conjugative elements1-3. Here we explore an understudied plasmid spread route among bacteria, mediated by intercellular membranous nanotube conduits4. We reveal that, unlike traditional donor-to-recipient delivery, Nanotube-dependent Plasmid exchange (NPex) operates bidirectionally, enabling both plasmid donation and, to a lesser extent, plasmid acquisition. By identifying a Bacillus subtilis natural isolate deficient in NPex, we discovered a prophage-encoded factor, YokF, that blocks plasmid transmission, chiefly acting within the donor cell to inhibit plasmid donation. YokF is a nuclease that localizes to the membrane, where it interacts with a nanotube component to selectively impede plasmid transfer through degradation. Importantly, YokF homologs from various bacterial species were found to exhibit anti-NPex activity. Given their prevalence in Gram-positive bacteria, we propose that YokF homologs represent a conserved family of NPex gatekeepers that restrict plasmid flow within bacterial communities.

microbiology↗

bioRxiv2025

PSD-95 drives binocular vision maturation critical for predation

Postsynaptic density protein 95 (PSD-95) is a signalling scaffold within the postsynaptic density of excitatory synapses which drives silent synapse maturation during critical periods (CP). Binocularity develops during visual CPs and matures before its closure. Despite lifelong critical period plasticity, PSD-95 knock-out (KO) mice were reported with only subtle sensory phenotypes as adults. To assess PSD-95s role in ethologically relevant binocular visual processing, we compared prey capture behaviour in PSD-95 KO and wild-type (WT) mice. KO mice were profoundly impaired in diverse epochs of predatory behaviour, but exhibited improved prey localisation under monocular conditions, reminiscent of impaired binocular integration. This was confirmed in an orientation discrimination task, where KO mice were impaired binocularly but performed monocularly like WT mice. Furthermore, binocular orientation discrimination was impaired after knocking down PSD-95 in primary visual cortex (V1), while a superior colliculus-specific knock-down had no effect. These results support an essential role of PSD-95 in binocular behaviour, which becomes evident under ethologically demanding behaviours.

neuroscience↗

bioRxiv2025

UCA1 lncRNA represses γ-globin expression by sequestering miR-148b, a key post-transcriptional regulator of BCL11A

Fetal hemoglobin (HbF; 2{gamma}2) reactivation is a promising strategy to ameliorate {beta}-hemoglobinopathies. However, limited understanding of {gamma}-globin (HBG1/2) regulation constrains development of therapeutic interventions. BCL11A, a key transcriptional repressor of {gamma}-globin, is central to HbF silencing during adult erythropoiesis. Here, we identify a new post-transcriptional regulatory mechanism involving the lncRNA-UCA1 and miR-148b that modulates BCL11A expression. Using UCA1 knockdown and overexpression strategies, combined with in vivo crosslinking and transcriptomic analyses, we demonstrate that UCA1 functions as a competing endogenous RNA (ceRNA), sequestering miR-148b and thereby attenuating its repressive effect on BCL11A. In the present study, we have elucidated the physiological significance of this interaction in adult erythroid cells, including CD34 HSPCs and HUDEP-2 cells, in which UCA1 depletion led to robust {gamma}-globin induction, a phenotype recapitulated by miR-148b overexpression. These findings uncover a previously unrecognized lncRNA-miRNA-mRNA regulatory axis and highlight the UCA1/miR-148b axis as a potential therapeutic target for HbF reactivation in {beta}-hemoglobinopathies.

genetics↗

bioRxiv2025

Reductive evolution of the DNA replication machinery in endosymbiotic fungi

The molecular machinery for replicating and repairing DNA accurately is critically important for life and highly conserved across the Tree of Life. Here we show that two major lineages of fungi, Glomeromycotina and Microsporidia, lost DNA polymerase complexes involved in replication and translesion synthesis. Catalytic and non-catalytic subunits of DNA polymerases are co-eliminated, consistent with their physical and functional interactions described in other eukaryotes. We detect lineage-specific variation in genome-wide mutation rates, showing that DNA polymerase gene losses correlate with increased genetic variation. We find that the Glomeraceae family of arbuscular mycorrhizal (AM) fungi has lived for [~]360 My without a leading strand replisome, raising the question of how these fungi can replicate DNA. We provide evidence that the cell cycle of Rhizophagus irregularis is active when in symbiosis with a host, but not without. This indicates a higher level of integration between AM fungi and plants than previously appreciated, and suggests the existence of a regulatory or functional contribution provided by a host to the fungal cell cycle. We propose that alternative modes of DNA replication and cell cycle provide mutational opportunities for fungal adaptation, and play roles in the evolution of endosymbioses.

evolutionary biology↗

bioRxiv2025

DRUMBEAT: Temporally resolved interpretable machine learning model for characterizing state transitions in protein dynamics

Conformational transitions are central to protein function, yet their mechanistic analysis remains challenging due to the multi-dimensionality and timescales underlying the molecular motions. While interpretable network models such as Bayesian networks have advanced the identification of key residue interactions in molecular dynamics (MD) data, they lack temporal resolution and cannot capture the sequence of events during transitions. Here, we introduce Dynamically Resolved Universal Model for BayEsiAn network Tracking or DRUMBEAT, a machine learning approach that combines a universal graph topology with sliding-window rescoring to generate interpretable, time-resolved maps of cooperative events in MD trajectories. Applying DRUMBEAT to the benchmark Fip35 WW domain folding trajectories from DE Shaw Research Group, we recover both major folding pathways and critical residues previously highlighted by experiment. Importantly, DRUMBEAT provides new insight in two ways: (1) uncover unknown protein features important for transition, and (2) dissect the order and timing of conformational changes, revealing the precise sequence of residue contact closures during individual folding events. Robustness analysis demonstrates that both the universal graph and time-resolved results are highly consistent across multiple sampling replicates. These findings establish DRUMBEAT as a scalable and interpretable machine learning framework for dissecting the dynamics of protein folding and other conformational transitions, offering a generalizable tool for the mechanistic study of biomolecular dynamics.

biochemistry↗

bioRxiv2025

Evaluating the Significance of Embedding-Based Protein Sequence Alignment with Clustering and Double Dynamic Programming for Remote Homology

Accurate detection of protein sequence homology is essential for understanding evolutionary relationships and predicting protein functions, particularly for detecting remote homology in the "twilight zone" (20-35% sequence similarity), where traditional sequence alignment methods often fail. Recent studies show that embeddings from protein language models (pLM) can improve remote homology detection over traditional methods. Alignment-based approaches, such as those combining pLMs with dynamic programming alignment, further improve performance but often suffer from noise in the resulting similarity matrices. To address this, we evaluate a newly developed embedding-based sequence alignment approach that refines residue-level embedding similarity using K-means clustering and double dynamic programming (DDP). We show that the incorporation of clustering and DDP contributes substantially to the improved performance in detecting remote homology. Experimental results demonstrate that our approach outperforms both traditional and state-of-the-art approaches based on pLMs on several benchmarks. Our study illustrates embedding-based alignment refined with clustering and DDP offers a powerful approach for identifying remote homology, with potential to evolve further as pLMs continue to advance.

bioinformatics↗

bioRxiv2025

An Atomic Force Microscopic analysis of Exosomes derived from Tumor Associated pluripotent Mesenchymal Stem Cells.

Mesenchymal stem cell (MSC)-derived exosomes are recognized as critical mediators within the tumor microenvironment (TME), exerting both pro- and anti-tumorigenic effects depending on contextual factors. These vesicles are also gaining attention for their potential as therapeutic vehicles in regenerative medicine and targeted drug delivery. However, the influence of the TME on the physical characteristics of MSC-derived exosomes remains poorly understood. In this study, we utilized Atomic Force Microscopy (AFM) to investigate the morphological and nanomechanical properties of MSC-derived exosomes under standard and TME-like conditions. AFM imaging in fluid mode preserved the native structure of exosomes and enabled high-resolution analysis of their topography, surface roughness, stiffness, adhesion, and deformation. AFM offers unique advantages in exosome research by enabling label-free, nanoscale analysis of vesicle properties in near-physiological conditions. The ability to detect such subtle but functionally significant changes highlights the relevance of AFM in exosome characterization and quality assessment. Our results revealed that exposure to the TME induces marked changes in exosomal membrane morphology and mechanical behavior, including increased surface heterogeneity, higher stiffness, and altered adhesive interactions. These biophysical alterations may reflect changes in membrane composition and protein or lipid cargo, potentially affecting exosome function, uptake, and therapeutic efficacy. Overall, our findings provide new insights into how the TME modulates MSC exosome biophysics and underscore the utility of AFM-based techniques for advancing the development of exosome-based diagnostics and therapies.

cancer biology↗

bioRxiv2025

Speed Vascular Patterns in the Spatial Navigation System

The hippocampal formation is central to spatial navigation, hosting neurons that encode position, direction, and speed. Yet, the brain-wide vascular dynamics supporting these processes remain poorly understood, especially during naturalistic behaviors. Here, we adapted functional ultrasound (fUS) imaging to examine how cerebral blood volume (CBV) changes relate to behavioral parameters in freely moving rats. High-resolution imaging of hippocampal-parahippocampal regions during open-field exploration reveals strong correlations between CBV dynamics and animal speed, with distinct regional activation patterns and temporal delays. Lagged general linear modeling uncovers information flow from the thalamus to parahippocampal regions, including the medial entorhinal cortex, and to hippocampal subfields (dentate gyrus, CA1-CA3), consistent with a hierarchical processing framework. The analysis also links CBV with angular head speed and the dorsal thalamus. Decoding analyses show that CBV signals not only encode speed precisely but also capture spatial features like proximity to walls and corners, even when univariate analyses do not. This decoding remains robust across animals, underscoring the universality of speed encoding in vascular dynamics. We also identify slow CBV oscillations in the hippocampus aligned with minute-scale speed fluctuations, suggesting a neurovascular signature of exploratory behavior. These findings reveal a hemodynamic signature of speed representation in the navigation system, arising from energy demands in a continuous attractor network model for path integration, where population activity and synaptic currents increase quadratically with animal speed as both peak firing rates and neuronal recruitment scale linearly with animal speed. Moreover, they highlight functional ultrasound imaging as a powerful approach for probing the hemodynamic basis of navigation.

neuroscience↗

bioRxiv2025

Auto-MedCalc: Automated Biomarkers Discovery and Risk Score Generation with AI Agents

Reproducible biomarker identification and transplant rejection risk prediction remain fundamental yet unsolved challenges in transplantation medicine. Traditional approaches rely on hypothesis-driven analyses and domain expertise, limiting scalability and generalizability across diverse populations. We introduce Transplant-Agents, a data-driven multi-agent AI framework integrating large language models (LLMs) with machine learning algorithms for automated biomarker identification and rejection risk prediction. Agents interact through structured, iterative dialogue governed by predefined rules and criteria, converging on optimal biomarker sets reproducible across multiple iterations. We evaluated three multicenter clinical trial transplant datasets from ImmPort, comprising 683 patients across kidney, liver, and heart transplant cohorts. Transplant-Agents achieve AUROC scores of 0.93, 0.88, and 0.88. Feature importance analysis further confirms the stability, interpretability and potential generalizability of identified biomarkers. This work demonstrates that AI-agent frameworks can reliably reproduce established transplant biomarkers while enabling transparent, validated, and standardized risk prediction pipelines.

bioinformatics↗

bioRxiv2025

Heterotypic Protein Interactions Modulate the Condensate Dynamics and Aggregation of α-Synuclein

Multicomponent biomolecular condensates formed by diverse multivalent proteins underlie numerous cellular processes, from ribosome biogenesis to stress response regulation, and have recently been implicated in disease-related protein aggregation. Understanding how heterotypic protein interactions modulate condensate dynamics and transitions to amyloid states remains a major challenge. Here, we combine fluorescence -based ensemble and single-molecule approaches, molecular simulations, and systematic domain deletions to investigate how charge patterning and domain structure influence co-condensation between -synuclein, a disordered neuronal protein linked to Parkinsons disease, and the SARS-CoV-2 nucleocapsid protein (NP), a structured viral RNA-binding protein. We find that co-condensation is driven by multivalent electrostatic interactions and occurs when the heterotypic affinity exceeds a threshold, resulting in restricted protein dynamics and altered condensate material properties. These changes promote the formation of dense amyloid fibrils, as confirmed by atomic force microscopy and fluorescence assays. Our results elucidate how heterotypic interactions within multiphasic condensates can modulate phase behavior and aggregation, providing insight into broader mechanisms linking condensate dysregulation with neurodegenerative diseases.

biophysics↗

bioRxiv2025

ImmuneFM: Pre-training Foundation Model from Cytometry Data for Immunology Research

Immunology is an essential field in the biomedicine domain, which plays an important role in oncology, vaccines, infection, etc. With the increasing amount of data available in immunology and artificial intelligence technique development, there is a need to develop data-driven AI methods in the field. However, the data from various immunology studies is very hard to integrate into an AI-ready dataset due to the lack of a standard. Moreover, independent immunology studies data lacks enough labels to train the supervised model. Motivated by these challenges, we curated a large-scale AI-ready cytometry dataset for immunology from the publicly available ImmPort portal. We design the framework to pre-train a foundation model, ImmuneFM, on the cytometry dataset. ImmuneFM can be applied to a wide range of downstream immunology diseases with fine-tuning on a limited number of labeled samples. The experiment results on eight downstream tasks demonstrate the superior performance of ImmuneFM compared to baseline deep learning and traditional methods.

immunology↗

bioRxiv2025

Benchmarking Large Language Models for Predictive Modeling in Biomedical Research With a Focus on Reproductive Health

Generative AI, particularly large language models (LLMs), is increasingly being used in computational biology to support code generation for data analysis. In this study, we evaluated the ability of LLMs to generate functional R and Python code for predictive modeling tasks, leveraging standardized molecular datasets from several recent DREAM (Dialogue for Reverse Engineering Assessments and Methods) Challenges focused on reproductive health. We assessed LLM performance across four predictive tasks derived from three DREAM challenges: gestational age regression from gene expression, gestational age regression from DNA methylation profiles, and classification of preterm birth and early preterm birth from microbiome data. LLMs were prompted with task descriptions, data locations, and target outcomes. LLM-generated code was then run to fit and apply prediction models and generate graphics, and they were ranked based on their success in completing the tasks and achieving strong test set performance. Among the eight LLMs tested, o3-mini-high, 4o, DeepseekR1 and Gemini 2.0 completed at least one task without error. Overall, R code generation was more successful (14/16 tasks) than Python (7/16), attributed to the utility of Bioconductor packages for querying Gene Expression Omnibus data. OpenAIs o3-mini-high outperformed others, completing 7/8 tasks. Test set performance of the top LLM matched or exceeded top-performing teams from the original DREAM challenges. These findings underscore the potential of LLMs to enhance exploratory analysis and democratize access to predictive modeling in omics by automating key components of analysis pipelines, and highlight the potential to increase research output when conducting analyses of standardized datasets from public repositories.

bioinformatics↗

bioRxiv2025

Recapitulating physiologically relevant oxygen levels and extracellular matrix remodeling in patient-derived tumor-immune tunable models reveal targeting opportunities for immunologically cold high-grade serous tumors

High-grade serous tumors are immunologically cold, characterized by limited immune cell infiltration and reduced clinical outcome, primarily due to hypoxia and extensive extracellular matrix remodeling that disrupt tumor-stromal-immune interactions. However, current experimental models fail to fully capture oxygen and matrix microenvironmental features, limiting progress in understanding tumor-immune dynamics and developing effective treatments. Here, we demonstrate that patient-derived tumor-immune tunable models, mimicking physiologically relevant oxygen levels and extracellular matrix remodeling, recapitulate the hypoxia-induced stromal/matrix dysregulation, which causes impaired immune infiltration, and enable dissecting targeted opportunities via TGF-{beta} signaling. The models integrate cancer cells co-cultured with cancer-associated fibroblasts and exposed to immune cells as multi-culture or challenged them to infiltrate into a 3D model bioengineered with autologous plasma from the matching patient or onto decellularized human ovaries. By bioengineering physiologically relevant oxygen levels of hypoxic tumors and physoxic ovaries, we uncovered that intratumoral hypoxia acts as a friend and a foe, causing hypoxia-induced stromal-driven impaired immune infiltration but enhancing the activation and cytotoxicity of CD8+ T cells. We also showed that targeting TGF-{beta} signaling reversed the hypoxia-induced stromal-driven impaired immune infiltration. These human-relevant tunable models may aid the development of targeted therapies to turn immunologically cold tumors into hot ones.

bioengineering↗