Biology subjects

Sundar, A.

Publications and source records attributed to Sundar, A..

7 recordsLinked to original sources

bioRxiv2026

Integrated Transcriptomic and CRISPR Dependency Analysis Prioritizes a CDK1-AURKB Mitotic Vulnerability Axis in Diffuse Intrinsic Pontine Glioma

Diffuse intrinsic pontine glioma (DIPG), now classified within diffuse midline glioma, H3K27-altered, remains a lethal pediatric brainstem tumor with limited therapeutic options. Here, we integrated public DIPG transcriptomic datasets, protein-protein interaction modeling, functional enrichment, immune deconvolution, survival analysis, and DepMap CRISPR dependency data to nominate candidate mitotic vulnerabilities. Differential expression analysis comparing 27 DIPG tumors with 6 brainstem low-grade glioma comparator samples identified a proliferative transcriptional program enriched for chromosome segregation, nuclear division, and cell-cycle pathways. Network analysis prioritized a compact mitotic hub module containing CDK1, AURKB, TOP2A, CDC20, CDCA8, and related G2/M regulators. CIBERSORT analysis of an independent DIPG cohort inferred low cytotoxic T-cell signal, consistent with an immune-cold phenotype, although immune-cell fractions require orthogonal validation. Survival analysis showed that neither inferred immune scores nor a composite mitotic hub score significantly stratified overall survival. DepMap CRISPR gene-effect data nominated CDK1, AURKB, TOP2A, and BIRC5 as candidate dependencies across brain tumor models. These findings provide a computational framework for prioritizing mitotic vulnerabilities in DIPG and support experimental validation in disease-relevant models.

bioinformatics↗

bioRxiv2026

Benchmarking Large Language Models for Predicting Therapeutic Antisense Oligonucleotide Efficacy

Antisense oligonucleotides (ASOs) are a promising class of therapeutic drugs that can target and modulate genes associated with various diseases. This study benchmarks Large Language Models (LLMs) for predicting ASO therapeutic efficacy through a two-stage approach: (1) molecular embedding-based fine-tuning using SMILES representations, and (2) prompt engineering with zero-shot and few-shot learning using DNA sequences with target gene information. We evaluated general-purpose models (GPT-3.5-Turbo, LLaMA2-7B, Galactica-6.7B) and chemistry-specific models (ChemBERTa, Molformer, BERT) across three datasets: PFRED (522 sequences), openASO (1708 sequences), and ASOptimizer (1267 sequences). DNA sequence inputs with target gene information outperformed SMILES representations. GPT-3.5-Turbo achieved R2 values of 0.6381 (PFRED) and 0.6340 (ASOptimizer) for few-shot prompting with k=3 examples. Code and datasets available at: https://github.com/asundar0128/IndependentStudy

bioinformatics↗

bioRxiv2025

A Computational Pipeline for Glioblastoma Vaccine Development: Integrating Novel Omics-Driven OIP5 Target Discovery to Create a Deep Learning-Based Immunogenicity Framework for Personalized Immunotherapy

This work introduces a modular, open-source computational pipeline for glioblastoma (GBM) vaccine design that integrates omics-based OIP5 target discovery with a deep learning framework for epitope immunogenicity prediction. Building upon conventional affinity-based predictors such as NetMHCpan, our Tumor Epitope Immunogenicity Pipeline (TEIP) incorporates biological, structural, and transcriptomic context to predict tumor-specific T-cell responses. Using curated immunogenic and non-immunogenic peptides from IEDB and GBM datasets, TEIP employs dual bidirectional LSTM encoders to represent peptide and HLA sequences, concatenated with auxiliary molecular features including proteasomal cleavage, TAP transport likelihood, gene expression, and mutation frequency. These features are fused into a probabilistic model that outputs peptide-specific immunogenicity scores and confidence estimates. Benchmark results across multiple HLA alleles demonstrate that TEIP outperforms NetMHCpan and MARIA, achieving an ROC-AUC of 0.89 and PR-AUC of 0.35. Feature importance analysis confirms that tumor-specific expression and peptide-MHC binding dominate predictive accuracy, validating the biological realism of the model. By combining mechanistic features with data-driven representation learning, TEIP enables fine-grained prioritization of neoantigens with translational potential. In proof-of-concept GBM studies, TEIP recovered known cancer-testis antigens such as OIP5, demonstrating its ability to identify clinically relevant vaccine targets. The pipeline is implemented using entirely open-access data and software to promote reproducibility and scalability. Collectively, TEIP provides a unified framework that connects multi-omics tumor characterization with deep learning-based antigen modeling, establishing a foundation for precision immunotherapy development in glioblastoma and beyond.

bioinformatics↗

bioRxiv2025

A secreted endosymbiont protein essential for colonizing host cells

ABSTRACTIntracellular bacterial symbioses have arisen myriad times in eukaryotes, with dozens known from insects alone1,2. Beginning with Buchnera, the obligate endosymbiont of aphids, genomes of endosymbionts have illuminated their evolutionary origins and metabolic contributions to hosts3,4. However, the mechanisms by which nonculturable endosymbionts enter host cells and suppress cellular immune processes have remained unknown. We show that an uncharacterized Buchnera protein, here designated SyeA, was present in the Buchnera ancestor, is secreted and homologous to secreted effectors of bacterial pathogens and is essential for Buchnera transmission. Buchnera is transmitted via expulsion from specialized maternal cells and uptake by embryos5. Using immunofluorescence microscopy, we found that SyeA levels peak upon colonization, accompanied by actin accumulation at the entry site. SyeA localizes outside the host-derived membrane and actin layer surrounding each Buchnera cell and colocalizes in host cytoplasm with Rho1, which regulates actin polymerization. syeA knockdown disrupts colonization and embryonic development and elevates lysosomal activity, leading to Buchnera destruction6. Our findings provide rare insight into how an anciently associated, mutualistic endosymbiont achieves its intracellular existence. SyeA is a vestige of pathogenic origins followed by evolution of increased host control and erosion of the original, more complex pathogenicity machinery.

microbiology↗

bioRxiv2025

Inhalable Perfluorocarbon Nanocapsules for RNA-based Immunotherapeutic Treatment of Lung Metastases

Inhalation RNA therapy offers to transform treatment of pulmonary diseases, yet mucus trapping, immune clearance, and navigation of heterogeneous lung tissue architecture still prevents RNA from reaching its target cells. Here, we develop perfluorocarbon (PFC) RNA nanocapsules that show negligible immune clearance, minimal inflammatory response, and efficient mucus transport, while passively homing to lung epithelial and tumor cells. After a single aerosolized dose in orthotopic lung metastasis model, more than 60% of tumor cells and most type II alveolar and bronchial epithelial cells internalized the nanocapsules, with observed pulmonary retention exceeding 48 h. The nanocapsule provoke negligible cytokine release, enabling repeated dosing. Treatment with therapeutic miR34-a suppresses metastatic outgrowth, potentiates anti-tumor immunity, and almost doubles median survival relative to control paclitaxel chemotherapy. By combining unique PFC disposition with RNA versatility, the delivery platform overcomes the main biological barriers for inhalable RNA medicines and opens a translatable path for treating diverse pulmonary diseases.

bioengineering↗

bioRxiv2023

Intracellular defensive symbiont is culturable and capable of transovarial, vertical transmission

Insects frequently form heritable associations with beneficial bacteria that are vertically transmitted from parent to offspring. Long term vertical transmission has repeatedly resulted in genome reduction and gene loss rendering many such bacteria incapable of independent culture. Among aphids, heritable endosymbionts often provide a wide range of context-specific benefits to their hosts. Although these associations have large impacts on host phenotypes, experimental approaches are often limited by an inability to independently cultivate these microbes. Here, we report the axenic culture of Candidatus Fukatsuia symbiotica strain WIR, a heritable bacterial endosymbiont of the pea aphid, Acyrthosiphon pisum. Whole genome sequencing revealed similar genomic features and high sequence similarity to previously described strains, suggesting the cultivation techniques used here may be applicable to Ca. F. symbiotica strains from distantly related aphids. Microinjection of the isolated strain into uninfected aphids revealed that it can reinfect developing embryos, and is maintained in subsequent generations via transovarial maternal transmission. Artificially infected aphids exhibit similar phenotypic and life history traits compared to native infections, including protective effects against an entomopathogenic Fusarium species. Overall, our results show that Ca. F. symbiotica may be a useful tool for experimentally probing the molecular mechanisms underlying heritable symbioses and antifungal defense in the pea aphid system. IMPORTANCEDiverse eukaryotic organisms form stable, symbiotic relationships with bacteria that provide benefits to their hosts. While these associations are often biologically important, they can be difficult to probe experimentally, because intimately host-associated bacteria are difficult to access within host tissues, and most cannot be cultured. This is especially true of the intracellular, maternally inherited bacteria associated with many insects, including aphids. Here, we demonstrate that a pea aphid-associated strain of the heritable endosymbiont, Candidatus Fukatsuia symbiotica, can be grown outside of its host using standard microbiology techniques, and can readily re-establish infection that is maintained across host generations. These artificial infections recapitulate the effects of native infections making this host-symbiont pair a useful experimental system. Using this system, we demonstrate that Ca. F. symbiotica infection reduces host fitness under benign conditions, but protects against a previously unreported fungal pathogen.

microbiology↗

bioRxiv2023

Intrinsic disorder in flaviviral capsid proteins and their role in pathogenesis

A high level of disorder in many viral proteins is a direct consequence of their small genomes, which makes interaction with multiple binding partners a necessity for infection and pathogenicity. A segment of the flaviviral capsid protein, also known as the molecular recognition feature (MoRF), undergoes a disorder-to-order transition upon binding to several protein partners. In order to understand their role in pathogenesis, the MORFs were identified and their homology was studied. Despite the lack of sequence similarities, the docking studies of Cs with the host proteins indicate conserved interactions involving MORFs across members of the phylogenetic subclades. Additionally, it was observed from the protein-protein networks that some MoRFs preferentially bind proteins that are involved in specialized functions such as ribosome biogenesis. The findings point to the importance of the MoRFs in the flaviviral life cycle, with important consequences for disease progression and suppression of the host immune system. Potentially, they might have impacted the way flaviviruses evolved to infect varied hosts using multiple vectors.

bioinformatics↗