Dhuvarakesh Karthikeyan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Dhuvarakesh Karthikeyan has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Alexander Rubinsteyn
Lineberger Comprehensive Cancer Center, The University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
5 shared publications - Andy Jinseok Lee
Lineberger Comprehensive Cancer Center, UNC Chapel Hill School of Medicine, Chapel Hill, NC, USA, 27514
3 shared publications - Benjamin G. Vincent
Lineberger Comprehensive Cancer Center, School of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA
3 shared publications - Misha Fini
Department of Microbiology and Immunology, UNC Chapel Hill School of Medicine, Chapel Hill, NC, USA, 27514
3 shared publications Benjamin Vincent
Personalized Immunotherapy Research Lab, UNC Chapel Hill
2 shared publications- Alex M. Jaeger
H. Lee Moffitt Cancer Center & Research Institute
1 shared publication
Latest Publications
- Open Access
Peptide:MHC Binding Stability Prediction Using Protein Language Models
A bstract: Peptide:MHC class I (pMHC-I) binding stability governs the persistence of antigenic complexes at the cell surface and plays a key role in facilitating downstream immunological signals such as antigen presentation, T-cell activation, and immunodominance. However, methods for in silico stability prediction remain underexplored relative to binding affinity prediction, in part because available half-life datasets are sparse and expensive to collect. Here, we perform a systematic reassessment of pMHC-I stability prediction using controlled, similarity-aware data splits and apply a recently introduced supervised transfer-learning strategy to MINT, an interaction-aware protein language model, pre-trained on binding affinity and fine-tuned for quantitative half-life prediction. We show that MINT improves stability prediction over standard ESM-2 representations and existing predictors, and that assay-conditioned recalibration corrects systematic shifts across experimental measurement modalities. Across eluted ligand, immunogenicity, and personalized neoantigen prioritization benchmarks, predicted stability provides signal beyond binding affinity, enriching for naturally presented and immunogenic peptides within affinity-filtered candidate sets. These results establish pMHC-I half-life as an orthogonal and transferable biophysical signal connecting peptide binding, surface presentation, and T-cell recognition, and provide a leakage-aware, assay-aware framework for future antigen-presentation modeling.
Open AccessPeptide:MHC Binding Stability Prediction Using Protein Language Models
A bstract: Peptide:MHC class I (pMHC-I) binding stability governs the persistence of antigenic complexes at the cell surface and plays a key role in facilitating downstream immunological signals such as antigen presentation, T-cell activation, and immunodominance. However, methods for in silico stability prediction remain underexplored relative to binding affinity prediction, in part because available half-life datasets are sparse and expensive to collect. Here, we perform a systematic reassessment of pMHC-I stability prediction using controlled, similarity-aware data splits and apply a recently introduced supervised transfer-learning strategy to MINT, an interaction-aware protein language model, pre-trained on binding affinity and fine-tuned for quantitative half-life prediction. We show that MINT improves stability prediction over standard ESM-2 representations and existing predictors, and that assay-conditioned recalibration corrects systematic shifts across experimental measurement modalities. Across eluted ligand, immunogenicity, and personalized neoantigen prioritization benchmarks, predicted stability provides signal beyond binding affinity, enriching for naturally presented and immunogenic peptides within affinity-filtered candidate sets. These results establish pMHC-I half-life as an orthogonal and transferable biophysical signal connecting peptide binding, surface presentation, and T-cell recognition, and provide a leakage-aware, assay-aware framework for future antigen-presentation modeling.
- Open Access
Immunodominance Hierarchy of Endogenous BBN963 Bladder Cancer Antigens Remains Stable Under Anti-PD1 and Anti-CTLA4 Immunotherapy
Immune checkpoint inhibition (ICI) is clinically active against multiple cancers, including urothelial cancer at the non-muscle invasive, muscle-invasive, and metastatic stages. Despite this, large numbers of patients experience disease progression and relapse after treatment with ICI-containing regimens. Tumor antigen-specific T cells are critical to ICI response, however few studies have evaluated the breadth and magnitude of tumor antigen-specific T cell responses with ICI therapy. In this study, we mapped the tumor antigen immunodominance hierarchy in the BBN963 model of murine basal-like bladder cancer for endogenous tumor neoantigens expressed physiologically. We used a high-throughput matrixed ELISpot assay to detect CD8 + T cell responses to predicted BBN963 tumor antigens derived from multiple mutational genomic sources. We found CD8+ T cell responses were directed against a subset of tumor antigens forming a stable and reproducible immunodominance hierarchy across individual mice. Treatment with anti-PD-1 or anti-CTLA-4 did not substantially reshape this hierarchy or broadly shift dominant responses to previously defined subdominant epitopes. Predicted peptide MHC binding stability and affinity was associated with antigen immunogenicity. Cancer-testis antigens, endogenous retroviral antigens, and SNV-derived tumor antigens that were immunogenic were found across tumor subclones. By diversifying the immunogenic antigen repertoire beyond SNVs, we achieved nearly 100% tumor subclone coverage, suggesting that broader antigen selection could help immunotherapy target more tumor subclones. In conclusion, this study supports the stability of the immunodominance hierarchy under ICI therapy and a role for broadening antigen discovery to multiple expressional sources in immunotherapy design.
Open AccessImmunodominance Hierarchy of Endogenous BBN963 Bladder Cancer Antigens Remains Stable Under Anti-PD1 and Anti-CTLA4 Immunotherapy
Immune checkpoint inhibition (ICI) is clinically active against multiple cancers, including urothelial cancer at the non-muscle invasive, muscle-invasive, and metastatic stages. Despite this, large numbers of patients experience disease progression and relapse after treatment with ICI-containing regimens. Tumor antigen-specific T cells are critical to ICI response, however few studies have evaluated the breadth and magnitude of tumor antigen-specific T cell responses with ICI therapy. In this study, we mapped the tumor antigen immunodominance hierarchy in the BBN963 model of murine basal-like bladder cancer for endogenous tumor neoantigens expressed physiologically. We used a high-throughput matrixed ELISpot assay to detect CD8 + T cell responses to predicted BBN963 tumor antigens derived from multiple mutational genomic sources. We found CD8+ T cell responses were directed against a subset of tumor antigens forming a stable and reproducible immunodominance hierarchy across individual mice. Treatment with anti-PD-1 or anti-CTLA-4 did not substantially reshape this hierarchy or broadly shift dominant responses to previously defined subdominant epitopes. Predicted peptide MHC binding stability and affinity was associated with antigen immunogenicity. Cancer-testis antigens, endogenous retroviral antigens, and SNV-derived tumor antigens that were immunogenic were found across tumor subclones. By diversifying the immunogenic antigen repertoire beyond SNVs, we achieved nearly 100% tumor subclone coverage, suggesting that broader antigen selection could help immunotherapy target more tumor subclones. In conclusion, this study supports the stability of the immunodominance hierarchy under ICI therapy and a role for broadening antigen discovery to multiple expressional sources in immunotherapy design.
- Open Access
- 8 September 2025
- Published by Springer Nature in Nature Machine Intelligence
- Vol. 7 (9) , 1494-1509
Conditional generation of real antigen-specific T cell receptor sequences
Despite recent advances in T cell receptor (TCR) engineering, designing functional TCRs against arbitrary targets remains challenging due to complex rules governing cross-reactivity and limited paired data. Here we present TCR-TRANSLATE, a sequence-to-sequence framework that adapts low-resource machine translation techniques to generate antigen-specific TCR sequences against unseen epitopes. By evaluating 12 model variants of the BART and T5 model architectures, we identified key factors affecting performance and utility, revealing discordances between these objectives. Our flagship model, TCRT5, outperforms existing approaches on computational benchmarks, prioritizing functionally relevant sequences at higher ranks. Most significantly, we experimentally validated a computationally designed TCR against Wilms’ tumour antigen, a therapeutically relevant target in leukaemia, excluded from our training and validation sets. Although the identified TCR shows cross-reactivity with pathogen-derived peptides, highlighting limitations in specificity, our work represents the successful computational design of a functional TCR construct against a non-viral epitope from the target sequence alone. Our findings establish a foundation for computational TCR design and reveal current limitations in data availability and methodology, providing a framework for accelerating personalized immunotherapy by reducing the search space for novel targets.
- 8 September 2025
- Published by Springer Nature in Nature Machine Intelligence
- Vol. 7 (9) , 1494-1509
Open AccessConditional generation of real antigen-specific T cell receptor sequences
Despite recent advances in T cell receptor (TCR) engineering, designing functional TCRs against arbitrary targets remains challenging due to complex rules governing cross-reactivity and limited paired data. Here we present TCR-TRANSLATE, a sequence-to-sequence framework that adapts low-resource machine translation techniques to generate antigen-specific TCR sequences against unseen epitopes. By evaluating 12 model variants of the BART and T5 model architectures, we identified key factors affecting performance and utility, revealing discordances between these objectives. Our flagship model, TCRT5, outperforms existing approaches on computational benchmarks, prioritizing functionally relevant sequences at higher ranks. Most significantly, we experimentally validated a computationally designed TCR against Wilms’ tumour antigen, a therapeutically relevant target in leukaemia, excluded from our training and validation sets. Although the identified TCR shows cross-reactivity with pathogen-derived peptides, highlighting limitations in specificity, our work represents the successful computational design of a functional TCR construct against a non-viral epitope from the target sequence alone. Our findings establish a foundation for computational TCR design and reveal current limitations in data availability and methodology, providing a framework for accelerating personalized immunotherapy by reducing the search space for novel targets.
- Open Access
TCR-TRANSLATE: Conditional Generation of Real Antigen Specific T-cell Receptor Sequences
The paradoxical nature of T-cell receptor (TCR) specificity, which requires both precise recognition and adequate coverage of antigenic peptide-MHCs (pMHCs), poses a fundamental challenge in immunology. Efforts at modeling this complex many-to-many mapping have been greatly impeded by a severe lack of experimental data. To address this, we present TCR-TRANSLATE, a novel framework that adapts low-resource machine translation techniques to the TCR:pMHC specificity domain. Here, we explore sequence-to-sequence (seq2seq) modeling with various training strategies, including semi-synthetic data augmentation and multi-task objectives to generate antigen specific TCR sequences for a given target of interest. We benchmark twelve model variants derived from the BART and T5 model architectures on a target-rich validation set of well-studied pMHCs, finding an optimal model, TCRT5, that generated validated antigen-specific CDR3β sequences for previously unseen antigens. While current limitations include a narrow validation set and a focus on the CDR3β loop, our approach demonstrates the potential of seq2seq models in rapidly generating antigen-specific TCR repertoires, offering a promising avenue for increasing throughput in precision immunotherapies. Our findings highlight both the capabilities and limitations of sequence-based conditional TCR design, emphasizing the need for experimental validation to bridge the gaps between predictions, metrics, and functional capacity.
Open AccessTCR-TRANSLATE: Conditional Generation of Real Antigen Specific T-cell Receptor Sequences
The paradoxical nature of T-cell receptor (TCR) specificity, which requires both precise recognition and adequate coverage of antigenic peptide-MHCs (pMHCs), poses a fundamental challenge in immunology. Efforts at modeling this complex many-to-many mapping have been greatly impeded by a severe lack of experimental data. To address this, we present TCR-TRANSLATE, a novel framework that adapts low-resource machine translation techniques to the TCR:pMHC specificity domain. Here, we explore sequence-to-sequence (seq2seq) modeling with various training strategies, including semi-synthetic data augmentation and multi-task objectives to generate antigen specific TCR sequences for a given target of interest. We benchmark twelve model variants derived from the BART and T5 model architectures on a target-rich validation set of well-studied pMHCs, finding an optimal model, TCRT5, that generated validated antigen-specific CDR3β sequences for previously unseen antigens. While current limitations include a narrow validation set and a focus on the CDR3β loop, our approach demonstrates the potential of seq2seq models in rapidly generating antigen-specific TCR repertoires, offering a promising avenue for increasing throughput in precision immunotherapies. Our findings highlight both the capabilities and limitations of sequence-based conditional TCR design, emphasizing the need for experimental validation to bridge the gaps between predictions, metrics, and functional capacity.
- Open Access
- 22 November 2023
- Published by Oxford University Press (OUP) in Briefings in Bioinformatics
- Vol. 25 (1)
ACE configurator for ELISpot: optimizing combinatorial design of pooled ELISpot assays with an epitope similarity model
The enzyme-linked immunosorbent spot (ELISpot) assay is a powerful in vitro immunoassay that enables cost-effective quantification of antigen-specific T-cell reactivity. It is used widely in the context of cancer and infectious diseases to validate the immunogenicity of predicted epitopes. While technological advances have kept pace with the demand for increased throughput, efforts to increase scale are bottlenecked by current assay design and deconvolution methods, which have remained largely unchanged. Current methods for designing pooled ELISpot experiments offer limited flexibility of assay parameters, lack support for high-throughput scenarios and do not consider peptide identity during pool assignment. We introduce the ACE Configurator for ELISpot (ACE) to address these gaps. ACE generates optimized peptide-pool assignments from highly customizable user inputs and handles the deconvolution of positive peptides using assay readouts. In this study, we present a novel sequence-aware pooling strategy, powered by a fine-tuned ESM-2 model that groups immunologically similar peptides, reducing the number of false positives and subsequent confirmatory assays compared to existing combinatorial approaches. To validate ACE’s performance on real-world datasets, we conducted a comprehensive benchmark study, contextualizing design choices with their impact on prediction quality. Our results demonstrate ACE’s capacity to further increase precision of identified immunogenic peptides, directly optimizing experimental efficiency. ACE is freely available as an executable with a graphical user interface and command-line interfaces at https://github.com/pirl-unc/ace.
- 22 November 2023
- Published by Oxford University Press (OUP) in Briefings in Bioinformatics
- Vol. 25 (1)
Open AccessACE configurator for ELISpot: optimizing combinatorial design of pooled ELISpot assays with an epitope similarity model
The enzyme-linked immunosorbent spot (ELISpot) assay is a powerful in vitro immunoassay that enables cost-effective quantification of antigen-specific T-cell reactivity. It is used widely in the context of cancer and infectious diseases to validate the immunogenicity of predicted epitopes. While technological advances have kept pace with the demand for increased throughput, efforts to increase scale are bottlenecked by current assay design and deconvolution methods, which have remained largely unchanged. Current methods for designing pooled ELISpot experiments offer limited flexibility of assay parameters, lack support for high-throughput scenarios and do not consider peptide identity during pool assignment. We introduce the ACE Configurator for ELISpot (ACE) to address these gaps. ACE generates optimized peptide-pool assignments from highly customizable user inputs and handles the deconvolution of positive peptides using assay readouts. In this study, we present a novel sequence-aware pooling strategy, powered by a fine-tuned ESM-2 model that groups immunologically similar peptides, reducing the number of false positives and subsequent confirmatory assays compared to existing combinatorial approaches. To validate ACE’s performance on real-world datasets, we conducted a comprehensive benchmark study, contextualizing design choices with their impact on prediction quality. Our results demonstrate ACE’s capacity to further increase precision of identified immunogenic peptides, directly optimizing experimental efficiency. ACE is freely available as an executable with a graphical user interface and command-line interfaces at https://github.com/pirl-unc/ace.
- Open Access
ACE Configurator for ELISpot (ACE): Optimizing Combinatorial Design of Pooled ELISpot Assays with an Epitope Similarity Model
The ELISpot assay is a powerful in vitro immunoassay that enables cost-effective quantification of antigen-specific T-cell reactivity. It is widely used in the context of cancer and infectious diseases to validate the immunogenicity of epitopes. While technological advances in hardware and software have kept pace with the need for increased throughput, assay design and deconvolution methodology have largely remained stagnant. Current methods for designing multiplexed ELISpot assays are restricted to preset configurations, lack support for high-throughput scenarios, and ignore peptide identity during pool assignment. We introduce the ACE Configurator for ELISpot (ACE) to address these gaps. ACE generates optimized peptide-pool assignments from highly customizable user inputs and handles positive peptide deconvolution using assay readouts. We present a novel sequence-aware pooling strategy, powered by a fine-tuned ESM-2 deep sequence model to identify immunologically similar peptides, reducing the number of false positives and subsequent confirmatory assays. To validate the performance of ACE using real-world datasets, we conducted a comprehensive benchmark study against various design heuristics, deconvolution methods, and experimental conditions, contextualizing design parameter choices with their impact on precision and number of total pools. Our results demonstrate ACE’s capacity to further increase precision of identified immunogenic peptides, maximizing experimental efficiency at the bench-side. ACE is freely available as an executable with a graphical user interface and command-line interfaces at https://github.com/pirl-unc/ace .
Open AccessACE Configurator for ELISpot (ACE): Optimizing Combinatorial Design of Pooled ELISpot Assays with an Epitope Similarity Model
The ELISpot assay is a powerful in vitro immunoassay that enables cost-effective quantification of antigen-specific T-cell reactivity. It is widely used in the context of cancer and infectious diseases to validate the immunogenicity of epitopes. While technological advances in hardware and software have kept pace with the need for increased throughput, assay design and deconvolution methodology have largely remained stagnant. Current methods for designing multiplexed ELISpot assays are restricted to preset configurations, lack support for high-throughput scenarios, and ignore peptide identity during pool assignment. We introduce the ACE Configurator for ELISpot (ACE) to address these gaps. ACE generates optimized peptide-pool assignments from highly customizable user inputs and handles positive peptide deconvolution using assay readouts. We present a novel sequence-aware pooling strategy, powered by a fine-tuned ESM-2 deep sequence model to identify immunologically similar peptides, reducing the number of false positives and subsequent confirmatory assays. To validate the performance of ACE using real-world datasets, we conducted a comprehensive benchmark study against various design heuristics, deconvolution methods, and experimental conditions, contextualizing design parameter choices with their impact on precision and number of total pools. Our results demonstrate ACE’s capacity to further increase precision of identified immunogenic peptides, maximizing experimental efficiency at the bench-side. ACE is freely available as an executable with a graphical user interface and command-line interfaces at https://github.com/pirl-unc/ace .
- Open Access
- 1 November 2020
- Published by Elsevier in Cell Reports Medicine
- Vol. 1 (8) , 100132
Developing Human Skin Contains Lymphocytes Demonstrating a Memory Signature
- 1 November 2020
- Published by Elsevier in Cell Reports Medicine
- Vol. 1 (8) , 100132
Open AccessDeveloping Human Skin Contains Lymphocytes Demonstrating a Memory Signature