Computational Immunology: Tailoring Tools to Tackle the T-Cell Triad

Whether you’re a seasoned computational scientist or an armchair immunologist, the growing focus on the T-cell triad — MHC, peptide & TCR — in both infectious and chronic diseases is necessitating easy access to complex datasets and analysis tools that can make sense of the human adaptive immune response.

In a recent webinar, DNAnexus research platforms expert Ben Busby joined Ankita Das, Head of Product at immune profiling company MIODx, to discuss ways in which the companies are working together to make it easier to interrogate and interpret TCR (T-cell receptor) data.

As Das explained, T-cell composition and activity are at the center of the immune response and key to tracking immune health. The composition of a person’s T-cell repertoire can vary depending upon factors like age, environment, genetics, infection, and lifestyle. When T-cells sense an infection, they undergo a phenomenon called clonal amplification, wherein a subset of the T-cell repertoire will amplify to orchestrate an immune response, whether that be killing off infected cells or recruiting B cells to generate antibodies. Interpreting clone activity can reveal important clues about immune health and insight that could potentially lead to biomarker and therapeutic discovery.

However, scientists wishing to undertake such research face several challenges. “TCR data is currently very siloed, and the architecture available for hosting and analyzing the immense data sets involved are often not scalable,” Das said. 

The MIODx team set out to overcome these challenges by creating ClonoMap™, a SaaSportal hosted on DNAnexus Titan™, in which TCR repertoire libraries can be stored, managed and analyzed in a secure environment. 

ClonoMapTM subscribers upload raw sequence data from TCR repertoire libraries into the portal and run  the ClonoMap™ Immune Profiler analysis to discover repertoire features. A second tool, ClonoMapTM Immune Insight, searches public datasets to help scientists see the new immune profiles in context and draw translational insights, such as biomarker identification. The MIODx team is now taking it a step further, applying machine learning to create personalized, immune health scores.

The ClonoMapTM suite of tools has already received nods for its innovation during the precisionFDA COVID-19 Precision Immunology App-a-thon. In the webinar, Das shared examples of how it was used to generate data and insights in COVID-19 and rheumatoid arthritis. In the case of COVID-19, for example, the Immune Profiler highlighted specific T Cell Receptor Beta Variable (TRBV) genes and CDR3 clonotypes at different frequencies in healthy individuals compared to COVID-19 recovered patients, providing TCRs for further investigation with respect to disease severity.

Holy grail of healthcare

Unravelling the immune response is the ‘holy grail of healthcare,’ Busby said. And in what may prove to be the decade of infectious disease research, Busby said he was proud to be able to provide tools to help scientists do so. 

“We want to make this size data accessible to everyone, and the DNAnexus platform really enables scientists and bioinformaticians to be more powerful,” Busby said. 

In addition to ClonoMap™, the iReceptor data discovery platform, curated by the AIRR (Adaptive Immune Receptor Repertoire) community, facilitates the curation, analysis and sharing of antibody/B-cell and T-cell receptor repertoires (AIRR-seq data) from multiple labs and institutions.

JupyterLab is another powerful tool that DNAnexus leverages for multi-omic cohort analysis and data exploration, and Busby recommended it for experts and armchair enthusiasts alike. He also noted that many DNAnexus-created Jupyter Notebooks are available, even to non-DNAnexus users. 

Other open source tools include Bioconductor, Bioconda and Docker. Each can be easily integrated into DNAnexus platforms, and the visualization capabilities of the DNAnexus system make the data obtained by them even more approachable, Busby said.  

You can watch the full webinar below.

How New Practices in Biomedical Research are Changing the Future of Healthcare

Biomedical Research Changing Healthcare

Five years ago, we were ushering in the ‘era of precision medicine,’ with researchers and clinicians alike embracing advances in genomics to help tailor treatments to individual patients, especially in cancer care. 

How has biomedical research and clinical care changed since then, and what does the future hold?

Precision medicine continues to garner attention, with its applications expanding into other areas, such as neuroscience, immunology, women’s health, and rare diseases. 

Patients are not only the focus of targeted therapies; they are also being placed front and center of many aspects of healthcare. 

Researchers are eager to incorporate diverse data from more representative patient populations, and registries from large-scale public sequencing efforts and patient advocacy groups are proving to be valuable resources in these endeavours.   

They are also more willing than ever to work together to accelerate discovery, thanks in part to the recent (and ongoing) global COVID-19 pandemic. 

How prepared are we to not only cope with these changes, but also harness their enormous potential? 

From Real-World Data to Real-World Evidence to Real-World Action

Advancements in technology supporting genomics, proteomics, metabolomics (and all the other ‘omics) have generated great insights into the biology of many diseases — and an enormous amount of information. Managing the vast ocean of data and fishing for answers within it are two of the primary challenges in precision medicine.

City of Hope is tackling the data challenge by creating a system that pools a wide variety of data from multiple sources in a way that can be easily accessed by bioinformaticians and physicians alike. 

Created with the help of DNAnexus, the system integrates DNA sequencing information with other data affiliated with the patient journey, from disease registries to pathologies, molecular characterization of the tumor, medical record data, and clinical trials information.

The POSEIDON platform has become more than just a static repository of data. It’s helped inform City of Hope’s unique in-house drug development. It’s helped place patients into clinical trials. It’s assisted tumor boards, where clinicians, researchers, and technical curators come together to make decisions on tricky cases. And it has led to new research ideas, new methods, and new translational projects. 

As precision medicine expands, so will the hunt for new biomarkers and the use of companion diagnostic tools. Laboratories will need an informatics environment that can flexibly scale to meet the demand for increased test volume. Cloud-based systems enable labs to optimize analysis pipelines for quality, speed, runtime, and cost, in a secure, compliant way.

Myriad Genetics uses the DNAnexus Titan platform to power its computationally intensive AI and machine learning methods. The company uses the technology for biomarker discovery, improvement of its current molecular diagnostic test portfolio, and disease risk prediction. 

The Apollo Platform and its Cohort Browser can also be used in pharmacogenomics to predict an individual’s risk to adverse drug reactions, another area that is likely to be in high demand as precision medicine becomes the norm.

Harnessing Rare Resources 

When researching a rare disease with many subtypes driven by diverse and distinct genetic alterations, data sharing is key. Samples acquired by a single institute, a single research initiative, or even a single nation may lack sufficient statistical power for genomic discovery and clinical correlative analysis.

St. Jude Children’s Research Hospital was an early adopter of cloud-based collaboration,  partnering with DNAnexus and Microsoft in 2018 to create a data-sharing ecosystem that has proved to be a model for harmonized genetic data across the pediatric cancer community. Since then, more than 1.25 petabytes of data have been incorporated into the St. Jude Cloud, and several research studies have been published about scientific discoveries made using the data.

The Muscular Dystrophy Association (MDA) is harnessing its patient registry to improve current and future patient care. Its neuroMuscular ObserVational Research (MOVR) data hub collects longitudinal data in seven neuromuscular disease indications, and a new visualization and analysis platform powered by DNAnexus is enabling 37 MDA Care Centers to easily access and analyze the information.

The MOVR Visualization and Reporting Platform (VRP) allows different levels of analysis, from overviews of disease progression and outcomes across sites, to in-depth dives into clinical parameters across large cohorts of neuromuscular patients. This level of correlative analyses could ultimately stimulate new drug, biologics and gene therapy discoveries. Exploration of deeply curated neuromuscular disease cohorts could also help in clinical trial design, by enabling clinical researchers to rapidly identify populations that meet specific clinical criteria.  

DNAnexus platforms are also being used by the Children’s Tumor Foundation to delve deeply into gene expression and transcriptome data to identify elusive therapeutic options for three forms of neurofibromatosis.

Going Global

Progress in science and medicine accelerates when researchers collaborate around responsibly shared datasets. As the complexity and scale of data increases, collaboration becomes more difficult to manage.

Tools developed on DNAnexus Apollo, such as the cloud-based UK Biobank Research Analysis Platform, have helped harmonize and democratize sequencing data by making it easily accessible to any scientist, from an individual field researcher accessing the database from her laptop, to pharma companies like Biogen, which is using the UKB information to rank candidate compounds in its drug portfolio as well as identify novel gene targets.  

The COVID-19 pandemic has underscored the importance of real-time data sharing, and an emerging focus on global pathogen surveillance will require even more companies to scale up their operations, from vaccine discovery and production, to rapid sample sequencing and diagnostics. 

The DNAnexus Apollo Platform was designed to seamlessly integrate and analyze diverse datasets, including multi-omic & clinical data, driving actionable insights in real time. We’ve compiled some tips for diagnostics businesses looking to scale their operations, as well as pointers for drug discovery companies

As we race towards a more interconnected, interpersonal future in healthcare, we need to ensure the industry is moving at the velocity of technological innovation. At DNAnexus, we’re proud to set the pace and provide solutions that allow all types of users to board the big data train.

How Multi-Omics is Changing Biomedical Research

Multi-Omics Changing Biomedical Research

Genomics, transcriptomics, proteomics, metagenomics… biology now seems to revolve around the “omics.” What exactly are they, why are they so crucial, and where is the field heading?

First off, a definition. When added to a biological word, “omics” refers to a global study of a system. So, for those of you who have always secretly wondered what the difference is between genetics and genomics: genetics is the study of individual genes, while genomics is the study of the entire genome. You’re welcome!

As the definition suggests, ‘omics research is interested in uncovering the often complex systems and molecular mechanisms underpinning diseases and other biological functions. 

Increasingly, scientists are combining ‘omics approaches for a more holistic, systems biology “the whole is greater than the sum of its parts” approach. 

Cancer provides a good example of why this approach is an attractive one. Oncogenic signalling is a multi-layered problem encompassing multiple molecular layers, from initial somatic mutations, to altered protein networks, to downstream activation of transcriptional programs. Combining genomics, proteomics, and transcriptomics can help elucidate activity along entire pathways. The simultaneous measurement of DNA and cell-surface protein, often called proteogenomic profiling, can also be used to characterize clonal diversity and evolution within tumors. And spatial ‘omics is providing yet another valuable layer of in situ insight. By using several tools in the ‘omics toolkit, researchers can piece together the most in-depth molecular pictures possible.

This multi-omics approach comes with challenges, however. Foremost among them is making sense of data that are in different formats and piecing them together to make meaningful conclusions. 

To make multi-omics analysis easier and more accessible to researchers, there must be well curated, publically accessible datasets that enable scientists to build upon already existing molecular knowledge. Scalable infrastructures to store and manage large quantities of data are also needed.

Ideally, there should also be interoperable tools that allow scientists to integrate data from different molecular layers, and to add their own data to study relationships between them. Even better: A streamlined process in which tools are moved into pipelines and accessible via interactive, intuitive graphical interfaces

Luckily, these are all areas in which DNAnexus excels.

DNAnexus Apollo™ is designed to handle the scale and scope of multi-omics research. It is data model agnostic, allowing for interrogation of any structured or unstructured data type.

Biologists can dive into thousands of phenotype fields and millions of variants in seconds, or query multi-omic characteristics to build cohorts for in-depth analysis, exploring associations between genomic and linked clinical data. Its built-in interactive data visualization tools and secure, cloud-based collaborative workspace makes it an even more attractive option for multi-disciplinary teams scattered around campus — or the globe. And it’s as flexible as your research.

The new multi-dimensional, multi-omics scientific queries may be complex, but the computational solutions used to harness them need not be. We aim to simplify the infrastructure needed to support these new approaches, so that their full potential can be realized.