Nomic and Parse Build a Multimodal, Time-Resolved Dataset for AI Models of Immune Response
Licensable foundational dataset combines 100 million single-cell RNA profiles with more than 11,000 quantitative protein profiles from healthy-donor perturbations, with leading biopharma partners already participating
[MONTREAL, October 29,2026] — Nomic Bio Inc., a proteomics company, and Parse Biosciences, a provider of single-cell sequencing and a QIAGEN company, today announced their partnership to generate Cascade, a large-scale multimodal, time-resolved dataset designed to establish a foundational reference for healthy human immune response. Built from controlled perturbations across healthy donors, Cascade pairs single-cell transcriptomic measurements with quantitative secretome proteomics to capture how immune cell states change and how secreted protein signals evolve over time. The dataset will support the development and evaluation of AI models of immune biology and provide a framework that can be extended to disease-specific questions.
A multimodal map of human immune response
As models of biology become more ambitious, they increasingly require experimental datasets that capture multiple molecular layers, span larger numbers of cells and perturbations, and follow biological responses over time. Cascade was designed around those requirements. Using PBMCs from 12 healthy donors, it will pair single-cell transcriptomic and quantitative secretome measurements across multiple timepoints, creating a time-resolved view of how at least 14 immune cell types respond to individual and combinatorial cytokine, chemokine, and drug perturbations. In total, the dataset will include 100 million single-cell RNA-seq profiles, more than 11,000 proteomic profiles, and over 150,000 perturbation measurements.
The time-resolved design captures biology that a single endpoint can miss, including early signals, transient responses, and delayed protein secretion. This allows researchers to follow how immune responses develop rather than observing only their final state, creating a stronger reference for modeling immune signaling and the effects of immune-modulating therapies.
The resulting dataset provides a common multimodal reference that researchers can build upon with additional perturbations, disease contexts, donors, and data modalities. Because the experimental framework is extensible, Cascade can serve as a starting point for adding new biological contexts while maintaining continuity for model development and evaluation.
"Many therapies aim to modulate cytokine signaling. To model those responses with AI, we need a strong reference for how healthy immune cells respond across perturbations they release. Cascade creates that foundation and a framework to build on for disease-specific questions."
Proteomics as a core data layer for AI biology
"Many therapies aim to modulate cytokine signaling,” said Milad Dagher, Chief Executive Officer of Nomic Bio. “To model those responses with AI, we need a strong reference for how healthy human immune cells respond across perturbations and time, including the proteins they release. Cascade creates that foundation and a framework to build on for disease-specific questions.”
Parse's GigaLab will perform single-cell RNA sequencing to capture how distinct immune cell types respond to each perturbation, while Nomic's Omni 1000 platform will profile more than 1,000 secreted proteins from the same samples. Nomic’s quantitative protein measurements are designed to remain comparable across studies, making Cascade interoperable with future datasets spanning new donors, perturbations, and disease contexts.
Linking the RNA and protein measurements by donor, perturbation, and time also creates opportunities for AI model development and evaluation. Researchers can, for example, test whether specific cellular responses predict downstream secreted-protein profiles, or hold out donors and perturbations to evaluate how well models generalize to unseen biology. These data can support model training, adaptation, benchmarking, and evaluation.
“Cascade was designed from the beginning as a multimodal dataset,” said Charlie Roco, Ph.D., Co-founder and Chief Technology Officer of Parse Biosciences. “The scale and structure of the study give researchers a common experimental framework for modeling immune response today and extending those models as new biological contexts and datasets are added.”
Industry leaders join as foundational partners
Cascade has already attracted foundational partners across biopharma and computational drug discovery, including GSK, Pfizer, Sanofi, Graph Therapeutics, and Turbine, reflecting demand for multimodal datasets that can support AI modeling, drug discovery, and deeper understanding of human immune response.
Data generation for Cascade is underway. The datasetwill be available for licensing, organizations interested in Cascade or working with Nomic on proteomic and multimodal datasets for AI and computational biology can learn more here.
Nomic and Parse Build a Multimodal, Time-Resolved Dataset for AI Models of Immune Response
Subscribe to Updates
Get the latest insights and updates delivered straight to your inbox weekly.
[MONTREAL, October 29,2026] — Nomic Bio Inc., a proteomics company, and Parse Biosciences, a provider of single-cell sequencing and a QIAGEN company, today announced their partnership to generate Cascade, a large-scale multimodal, time-resolved dataset designed to establish a foundational reference for healthy human immune response. Built from controlled perturbations across healthy donors, Cascade pairs single-cell transcriptomic measurements with quantitative secretome proteomics to capture how immune cell states change and how secreted protein signals evolve over time. The dataset will support the development and evaluation of AI models of immune biology and provide a framework that can be extended to disease-specific questions.
A multimodal map of human immune response
As models of biology become more ambitious, they increasingly require experimental datasets that capture multiple molecular layers, span larger numbers of cells and perturbations, and follow biological responses over time. Cascade was designed around those requirements. Using PBMCs from 12 healthy donors, it will pair single-cell transcriptomic and quantitative secretome measurements across multiple timepoints, creating a time-resolved view of how at least 14 immune cell types respond to individual and combinatorial cytokine, chemokine, and drug perturbations. In total, the dataset will include 100 million single-cell RNA-seq profiles, more than 11,000 proteomic profiles, and over 150,000 perturbation measurements.
The time-resolved design captures biology that a single endpoint can miss, including early signals, transient responses, and delayed protein secretion. This allows researchers to follow how immune responses develop rather than observing only their final state, creating a stronger reference for modeling immune signaling and the effects of immune-modulating therapies.
The resulting dataset provides a common multimodal reference that researchers can build upon with additional perturbations, disease contexts, donors, and data modalities. Because the experimental framework is extensible, Cascade can serve as a starting point for adding new biological contexts while maintaining continuity for model development and evaluation.
"Many therapies aim to modulate cytokine signaling. To model those responses with AI, we need a strong reference for how healthy immune cells respond across perturbations they release. Cascade creates that foundation and a framework to build on for disease-specific questions."
Proteomics as a core data layer for AI biology
"Many therapies aim to modulate cytokine signaling,” said Milad Dagher, Chief Executive Officer of Nomic Bio. “To model those responses with AI, we need a strong reference for how healthy human immune cells respond across perturbations and time, including the proteins they release. Cascade creates that foundation and a framework to build on for disease-specific questions.”
Parse's GigaLab will perform single-cell RNA sequencing to capture how distinct immune cell types respond to each perturbation, while Nomic's Omni 1000 platform will profile more than 1,000 secreted proteins from the same samples. Nomic’s quantitative protein measurements are designed to remain comparable across studies, making Cascade interoperable with future datasets spanning new donors, perturbations, and disease contexts.
Linking the RNA and protein measurements by donor, perturbation, and time also creates opportunities for AI model development and evaluation. Researchers can, for example, test whether specific cellular responses predict downstream secreted-protein profiles, or hold out donors and perturbations to evaluate how well models generalize to unseen biology. These data can support model training, adaptation, benchmarking, and evaluation.
“Cascade was designed from the beginning as a multimodal dataset,” said Charlie Roco, Ph.D., Co-founder and Chief Technology Officer of Parse Biosciences. “The scale and structure of the study give researchers a common experimental framework for modeling immune response today and extending those models as new biological contexts and datasets are added.”
Industry leaders join as foundational partners
Cascade has already attracted foundational partners across biopharma and computational drug discovery, including GSK, Pfizer, Sanofi, Graph Therapeutics, and Turbine, reflecting demand for multimodal datasets that can support AI modeling, drug discovery, and deeper understanding of human immune response.
Data generation for Cascade is underway. The datasetwill be available for licensing, organizations interested in Cascade or working with Nomic on proteomic and multimodal datasets for AI and computational biology can learn more here.
[MONTREAL, October 29,2026] — Nomic Bio Inc., a proteomics company, and Parse Biosciences, a provider of single-cell sequencing and a QIAGEN company, today announced their partnership to generate Cascade, a large-scale multimodal, time-resolved dataset designed to establish a foundational reference for healthy human immune response. Built from controlled perturbations across healthy donors, Cascade pairs single-cell transcriptomic measurements with quantitative secretome proteomics to capture how immune cell states change and how secreted protein signals evolve over time. The dataset will support the development and evaluation of AI models of immune biology and provide a framework that can be extended to disease-specific questions.
A multimodal map of human immune response
As models of biology become more ambitious, they increasingly require experimental datasets that capture multiple molecular layers, span larger numbers of cells and perturbations, and follow biological responses over time. Cascade was designed around those requirements. Using PBMCs from 12 healthy donors, it will pair single-cell transcriptomic and quantitative secretome measurements across multiple timepoints, creating a time-resolved view of how at least 14 immune cell types respond to individual and combinatorial cytokine, chemokine, and drug perturbations. In total, the dataset will include 100 million single-cell RNA-seq profiles, more than 11,000 proteomic profiles, and over 150,000 perturbation measurements.
The time-resolved design captures biology that a single endpoint can miss, including early signals, transient responses, and delayed protein secretion. This allows researchers to follow how immune responses develop rather than observing only their final state, creating a stronger reference for modeling immune signaling and the effects of immune-modulating therapies.
The resulting dataset provides a common multimodal reference that researchers can build upon with additional perturbations, disease contexts, donors, and data modalities. Because the experimental framework is extensible, Cascade can serve as a starting point for adding new biological contexts while maintaining continuity for model development and evaluation.
"Many therapies aim to modulate cytokine signaling. To model those responses with AI, we need a strong reference for how healthy immune cells respond across perturbations they release. Cascade creates that foundation and a framework to build on for disease-specific questions."
Proteomics as a core data layer for AI biology
"Many therapies aim to modulate cytokine signaling,” said Milad Dagher, Chief Executive Officer of Nomic Bio. “To model those responses with AI, we need a strong reference for how healthy human immune cells respond across perturbations and time, including the proteins they release. Cascade creates that foundation and a framework to build on for disease-specific questions.”
Parse's GigaLab will perform single-cell RNA sequencing to capture how distinct immune cell types respond to each perturbation, while Nomic's Omni 1000 platform will profile more than 1,000 secreted proteins from the same samples. Nomic’s quantitative protein measurements are designed to remain comparable across studies, making Cascade interoperable with future datasets spanning new donors, perturbations, and disease contexts.
Linking the RNA and protein measurements by donor, perturbation, and time also creates opportunities for AI model development and evaluation. Researchers can, for example, test whether specific cellular responses predict downstream secreted-protein profiles, or hold out donors and perturbations to evaluate how well models generalize to unseen biology. These data can support model training, adaptation, benchmarking, and evaluation.
“Cascade was designed from the beginning as a multimodal dataset,” said Charlie Roco, Ph.D., Co-founder and Chief Technology Officer of Parse Biosciences. “The scale and structure of the study give researchers a common experimental framework for modeling immune response today and extending those models as new biological contexts and datasets are added.”
Industry leaders join as foundational partners
Cascade has already attracted foundational partners across biopharma and computational drug discovery, including GSK, Pfizer, Sanofi, Graph Therapeutics, and Turbine, reflecting demand for multimodal datasets that can support AI modeling, drug discovery, and deeper understanding of human immune response.
Data generation for Cascade is underway. The datasetwill be available for licensing, organizations interested in Cascade or working with Nomic on proteomic and multimodal datasets for AI and computational biology can learn more here.
Nomic and Parse Build a Multimodal, Time-Resolved Dataset for AI Models of Immune Response
[MONTREAL, October 29,2026] — Nomic Bio Inc., a proteomics company, and Parse Biosciences, a provider of single-cell sequencing and a QIAGEN company, today announced their partnership to generate Cascade, a large-scale multimodal, time-resolved dataset designed to establish a foundational reference for healthy human immune response. Built from controlled perturbations across healthy donors, Cascade pairs single-cell transcriptomic measurements with quantitative secretome proteomics to capture how immune cell states change and how secreted protein signals evolve over time. The dataset will support the development and evaluation of AI models of immune biology and provide a framework that can be extended to disease-specific questions.
A multimodal map of human immune response
As models of biology become more ambitious, they increasingly require experimental datasets that capture multiple molecular layers, span larger numbers of cells and perturbations, and follow biological responses over time. Cascade was designed around those requirements. Using PBMCs from 12 healthy donors, it will pair single-cell transcriptomic and quantitative secretome measurements across multiple timepoints, creating a time-resolved view of how at least 14 immune cell types respond to individual and combinatorial cytokine, chemokine, and drug perturbations. In total, the dataset will include 100 million single-cell RNA-seq profiles, more than 11,000 proteomic profiles, and over 150,000 perturbation measurements.
The time-resolved design captures biology that a single endpoint can miss, including early signals, transient responses, and delayed protein secretion. This allows researchers to follow how immune responses develop rather than observing only their final state, creating a stronger reference for modeling immune signaling and the effects of immune-modulating therapies.
The resulting dataset provides a common multimodal reference that researchers can build upon with additional perturbations, disease contexts, donors, and data modalities. Because the experimental framework is extensible, Cascade can serve as a starting point for adding new biological contexts while maintaining continuity for model development and evaluation.
"Many therapies aim to modulate cytokine signaling. To model those responses with AI, we need a strong reference for how healthy immune cells respond across perturbations they release. Cascade creates that foundation and a framework to build on for disease-specific questions."
Proteomics as a core data layer for AI biology
"Many therapies aim to modulate cytokine signaling,” said Milad Dagher, Chief Executive Officer of Nomic Bio. “To model those responses with AI, we need a strong reference for how healthy human immune cells respond across perturbations and time, including the proteins they release. Cascade creates that foundation and a framework to build on for disease-specific questions.”
Parse's GigaLab will perform single-cell RNA sequencing to capture how distinct immune cell types respond to each perturbation, while Nomic's Omni 1000 platform will profile more than 1,000 secreted proteins from the same samples. Nomic’s quantitative protein measurements are designed to remain comparable across studies, making Cascade interoperable with future datasets spanning new donors, perturbations, and disease contexts.
Linking the RNA and protein measurements by donor, perturbation, and time also creates opportunities for AI model development and evaluation. Researchers can, for example, test whether specific cellular responses predict downstream secreted-protein profiles, or hold out donors and perturbations to evaluate how well models generalize to unseen biology. These data can support model training, adaptation, benchmarking, and evaluation.
“Cascade was designed from the beginning as a multimodal dataset,” said Charlie Roco, Ph.D., Co-founder and Chief Technology Officer of Parse Biosciences. “The scale and structure of the study give researchers a common experimental framework for modeling immune response today and extending those models as new biological contexts and datasets are added.”
Industry leaders join as foundational partners
Cascade has already attracted foundational partners across biopharma and computational drug discovery, including GSK, Pfizer, Sanofi, Graph Therapeutics, and Turbine, reflecting demand for multimodal datasets that can support AI modeling, drug discovery, and deeper understanding of human immune response.
Data generation for Cascade is underway. The datasetwill be available for licensing, organizations interested in Cascade or working with Nomic on proteomic and multimodal datasets for AI and computational biology can learn more here.
Access Resource
Nomic and Parse Build a Multimodal, Time-Resolved Dataset for AI Models of Immune Response
[MONTREAL, October 29,2026] — Nomic Bio Inc., a proteomics company, and Parse Biosciences, a provider of single-cell sequencing and a QIAGEN company, today announced their partnership to generate Cascade, a large-scale multimodal, time-resolved dataset designed to establish a foundational reference for healthy human immune response. Built from controlled perturbations across healthy donors, Cascade pairs single-cell transcriptomic measurements with quantitative secretome proteomics to capture how immune cell states change and how secreted protein signals evolve over time. The dataset will support the development and evaluation of AI models of immune biology and provide a framework that can be extended to disease-specific questions.
A multimodal map of human immune response
As models of biology become more ambitious, they increasingly require experimental datasets that capture multiple molecular layers, span larger numbers of cells and perturbations, and follow biological responses over time. Cascade was designed around those requirements. Using PBMCs from 12 healthy donors, it will pair single-cell transcriptomic and quantitative secretome measurements across multiple timepoints, creating a time-resolved view of how at least 14 immune cell types respond to individual and combinatorial cytokine, chemokine, and drug perturbations. In total, the dataset will include 100 million single-cell RNA-seq profiles, more than 11,000 proteomic profiles, and over 150,000 perturbation measurements.
The time-resolved design captures biology that a single endpoint can miss, including early signals, transient responses, and delayed protein secretion. This allows researchers to follow how immune responses develop rather than observing only their final state, creating a stronger reference for modeling immune signaling and the effects of immune-modulating therapies.
The resulting dataset provides a common multimodal reference that researchers can build upon with additional perturbations, disease contexts, donors, and data modalities. Because the experimental framework is extensible, Cascade can serve as a starting point for adding new biological contexts while maintaining continuity for model development and evaluation.
"Many therapies aim to modulate cytokine signaling. To model those responses with AI, we need a strong reference for how healthy immune cells respond across perturbations they release. Cascade creates that foundation and a framework to build on for disease-specific questions."
Proteomics as a core data layer for AI biology
"Many therapies aim to modulate cytokine signaling,” said Milad Dagher, Chief Executive Officer of Nomic Bio. “To model those responses with AI, we need a strong reference for how healthy human immune cells respond across perturbations and time, including the proteins they release. Cascade creates that foundation and a framework to build on for disease-specific questions.”
Parse's GigaLab will perform single-cell RNA sequencing to capture how distinct immune cell types respond to each perturbation, while Nomic's Omni 1000 platform will profile more than 1,000 secreted proteins from the same samples. Nomic’s quantitative protein measurements are designed to remain comparable across studies, making Cascade interoperable with future datasets spanning new donors, perturbations, and disease contexts.
Linking the RNA and protein measurements by donor, perturbation, and time also creates opportunities for AI model development and evaluation. Researchers can, for example, test whether specific cellular responses predict downstream secreted-protein profiles, or hold out donors and perturbations to evaluate how well models generalize to unseen biology. These data can support model training, adaptation, benchmarking, and evaluation.
“Cascade was designed from the beginning as a multimodal dataset,” said Charlie Roco, Ph.D., Co-founder and Chief Technology Officer of Parse Biosciences. “The scale and structure of the study give researchers a common experimental framework for modeling immune response today and extending those models as new biological contexts and datasets are added.”
Industry leaders join as foundational partners
Cascade has already attracted foundational partners across biopharma and computational drug discovery, including GSK, Pfizer, Sanofi, Graph Therapeutics, and Turbine, reflecting demand for multimodal datasets that can support AI modeling, drug discovery, and deeper understanding of human immune response.
Data generation for Cascade is underway. The datasetwill be available for licensing, organizations interested in Cascade or working with Nomic on proteomic and multimodal datasets for AI and computational biology can learn more here.
Download resource
Get the latest insights and updates delivered straight to your inbox weekly.
Nomic and Parse Build a Multimodal, Time-Resolved Dataset for AI Models of Immune Response
Get the latest insights and updates delivered straight to your inbox weekly.
[MONTREAL, October 29,2026] — Nomic Bio Inc., a proteomics company, and Parse Biosciences, a provider of single-cell sequencing and a QIAGEN company, today announced their partnership to generate Cascade, a large-scale multimodal, time-resolved dataset designed to establish a foundational reference for healthy human immune response. Built from controlled perturbations across healthy donors, Cascade pairs single-cell transcriptomic measurements with quantitative secretome proteomics to capture how immune cell states change and how secreted protein signals evolve over time. The dataset will support the development and evaluation of AI models of immune biology and provide a framework that can be extended to disease-specific questions.
A multimodal map of human immune response
As models of biology become more ambitious, they increasingly require experimental datasets that capture multiple molecular layers, span larger numbers of cells and perturbations, and follow biological responses over time. Cascade was designed around those requirements. Using PBMCs from 12 healthy donors, it will pair single-cell transcriptomic and quantitative secretome measurements across multiple timepoints, creating a time-resolved view of how at least 14 immune cell types respond to individual and combinatorial cytokine, chemokine, and drug perturbations. In total, the dataset will include 100 million single-cell RNA-seq profiles, more than 11,000 proteomic profiles, and over 150,000 perturbation measurements.
The time-resolved design captures biology that a single endpoint can miss, including early signals, transient responses, and delayed protein secretion. This allows researchers to follow how immune responses develop rather than observing only their final state, creating a stronger reference for modeling immune signaling and the effects of immune-modulating therapies.
The resulting dataset provides a common multimodal reference that researchers can build upon with additional perturbations, disease contexts, donors, and data modalities. Because the experimental framework is extensible, Cascade can serve as a starting point for adding new biological contexts while maintaining continuity for model development and evaluation.
"Many therapies aim to modulate cytokine signaling. To model those responses with AI, we need a strong reference for how healthy immune cells respond across perturbations they release. Cascade creates that foundation and a framework to build on for disease-specific questions."
Proteomics as a core data layer for AI biology
"Many therapies aim to modulate cytokine signaling,” said Milad Dagher, Chief Executive Officer of Nomic Bio. “To model those responses with AI, we need a strong reference for how healthy human immune cells respond across perturbations and time, including the proteins they release. Cascade creates that foundation and a framework to build on for disease-specific questions.”
Parse's GigaLab will perform single-cell RNA sequencing to capture how distinct immune cell types respond to each perturbation, while Nomic's Omni 1000 platform will profile more than 1,000 secreted proteins from the same samples. Nomic’s quantitative protein measurements are designed to remain comparable across studies, making Cascade interoperable with future datasets spanning new donors, perturbations, and disease contexts.
Linking the RNA and protein measurements by donor, perturbation, and time also creates opportunities for AI model development and evaluation. Researchers can, for example, test whether specific cellular responses predict downstream secreted-protein profiles, or hold out donors and perturbations to evaluate how well models generalize to unseen biology. These data can support model training, adaptation, benchmarking, and evaluation.
“Cascade was designed from the beginning as a multimodal dataset,” said Charlie Roco, Ph.D., Co-founder and Chief Technology Officer of Parse Biosciences. “The scale and structure of the study give researchers a common experimental framework for modeling immune response today and extending those models as new biological contexts and datasets are added.”
Industry leaders join as foundational partners
Cascade has already attracted foundational partners across biopharma and computational drug discovery, including GSK, Pfizer, Sanofi, Graph Therapeutics, and Turbine, reflecting demand for multimodal datasets that can support AI modeling, drug discovery, and deeper understanding of human immune response.
Data generation for Cascade is underway. The datasetwill be available for licensing, organizations interested in Cascade or working with Nomic on proteomic and multimodal datasets for AI and computational biology can learn more here.