Our approach to Immunology research
To truly understand the human immune system, we must look at it through many different lenses and at multiple time points. Practically speaking, that requires integrating several types of data to get a more complete picture of the roles different immune cells play in different contexts, and where and how dysregulation occurs in disease. To achieve that integration, the Allen Institute for Immunology has built a standardized research pipeline with several core assays to analyze blood and tissue samples from study volunteers who are either healthy or have disease over a period of two years.
Consistency and scalability are crucial for longitudinal studies. To meet these targets, the pipeline integrates a high throughput fluorescence-based cell counter, Nexcelom Cellaca MX, and liquid handling robotics for cell transfer steps in both flow cytometry and single cell genomics, the Integra VIAFLO 96.
Our pipeline integrates data collected from: flow cytometry, single-cell RNA sequencing, single-cell ATAC-seq, plasma proteomics, and clinical data. Viewing all these data types together creates a holistic view of the immune system and provides insights to the mechanisms that allow the immune system to appropriately fight off pathogens or inappropriately become dysregulated in immune diseases. Additional custom approaches are tailored to each disease state and to drive our hypothesis driven research.
Learn more about our technologies, pipelines, tools and data
Flow cytometry is a combination of fluorescence microscopy and a fluidics system that can simultaneously measure up to 40 proteins on each cell, at a rate of up to 50,000 cells per second. This technology has become a mainstay of modern immunology research due to its ability to measure the proportions of different types of immune cells present in a blood sample and capture many details about each individual cell including its activation state and function. This makes it a powerful tool for investigating the immune system in diseases such as cancer and autoimmunity.
The Allen Institute for Immunology uses both conventional and spectral flow cytometers but has developed particular expertise in high-throughput, high-dimensional spectral cytometry approaches. Our basic approach to most samples includes analysis of ~1.6 million immune cells using 4 different flow cytometry “panels”, each panel measuring 25-30 different proteins on every cell. The output is high quality longitudinal flow cytometry data over multiple batches. Recording this high number of cells improves the level of detection and statistical significance for rare immune cell types.
We also use high efficiency cell sorting to isolate various immune cell types for downstream studies including single cell technologies in the lab. We have integrated flow cytometry and cell sorting to complement our single-cell platforms, such as single-cell genomics and RNA sequencing, expanding our ability to look deeper into cellular processes and changes in the immune system with disease.
Single-cell sequencing technologies
The immune system is composed of many different cell types that also exist in different cell states depending on their environments. We use a variety of single-cell sequencing techniques, including some developed in-house, to measure the cell state. Single-cell RNA sequencing in combination with measurement of proteins by CITE-seq provides a detailed transcriptomic and protein analysis at the single-cell level, which is essential to follow the immune system’s response to pathogens in mounting an antigen specific response. CITE-seq allows vertical data integration with flow cytometry by evaluating the same proteins in both platforms. We have further developed the concept of cell hashing to simultaneously increase data quality and throughput for scRNA-seq. This has enabled us to profile the transcriptional state of millions of cells covering a wide variety of human immune states, including healthy, diseased and responses to drug treatments.
Single-cell ATAC sequencing maps open chromatin regions across the genome to identify DNA binding sites for key transcription factors that drive the transcriptional and phenotypic program of the cell as it responds to being stimulated. We have developed a unique combined method that combines all three assay types to simultaneously measure the RNA transcriptome, chromatin accessibility, and key marker proteins on the same single cells. This multi-omic, single cell approach that we call TEA-seq provides a complete picture of cell signaling inputs through cell surface markers, transcription factor signaling, and gene output at the RNA level.
The huge amount of molecular data generated from our experimental pipeline and the human metadata collected in our clinical studies provides us and our collaborators the opportunity to generate new insights into the immune system in health and disease. However, the size and complexity of these datasets brings new challenges. The Computational Biology team is utilizing established methods and building new approaches to manipulate and analyze large datasets to understand the human immune system. Our goal is to combine our large in-house data with published datasets to define gene regulatory networks, protein-cell interactions, cell-cell interactions, and signaling cascades that lie at the heart of immune-related disorders.
Human Immune System Explorer
We have developed the Human Immune System Explorer, a unique data storage, access, and research platform to gather, quality control, curate and compare data sets in a collaborative, traceable, cloud environment. Currently, the Human Immune System Explorer is used by Allen Institute for Immunology researchers and our research collaborators. Eventually, some parts of the data portal will be open to all.
Automated analysis pipelines provide consistent, fast and accurate analysis of flow cytometry and single cell data. Visualization tools provide the first opportunity to explore these results, and built-in integrated development environments enable computational scientists to collaborate on data analysis projects using established or novel computational packages. Human metadata including demographics, blood test results (complete blood counts, electrolyte levels, etc.), survey responses, and patient symptoms is overlaid to aid the interpretation of experimental data.
To facilitate collaboration, a variety of methods are available to easily share data and tools among the scientific teams. All data manipulations are automatically recorded, enabling tracing of the provenance of a result, verification, and reproducibility.
The Human Immune System Explorer is a cloud-based multi-tenant environment that enables authenticated and authorized access to research scientists. Its distributed architecture is implemented using a variety of best-in-class languages, data stores, frameworks, and CI/CD methodologies, including GoLang, React, R, Python, MongoDB, SQL, Google Cloud, Docker, Kubernetes, Argo, Terraform, Atlantis, and Harness.