Time Period: On going
Working closely with NCSA’s Healthcare Innovation Program Office, the Visual Analytics group is part of an ongoing collaboration with researchers at the NIH National Institute of Allergy and Infectious Disease, Institute for Systems Biology and Illinois Department of Statistics. This interdisciplinary team is taking an “all-inclusive” approach in studying the disease pathogenesis of influenzae and, more recently, COVID-19. Their goal is to understand the beginning-to-end biological responses to viral attack from the standpoint of multiple key players, including the invading virus, defending host cell, and overall system (e.g., human body). With comprehensive studies come vast amounts of data, and not all data that is statistically significant is biologically relevant. The team works together to pull out the pertinent information from the data and present them in a clear and simple way. Equipped with experts from biology and computer science and a certain curiosity in one another’s disciplines, the team aptly integrates the appropriate scientific contexts into advanced statistical and machine learning methods in order to accurately interpret time series data from ongoing clinical studies. The team’s publications thus far have provided insight on why and how viruses and immune cells behave, interact with each other and lead to the manifestation of clinical symptoms. Ultimately, they hope their findings will help identify actionable targets for new therapies for influenzae and COVID-19.
One of the goals for this project is to identify biomarkers in the blood that can predict factors such as viral shedding (the primary factor that determines contagiousness), disease severity, or risk for developing life-threatening complications in a patient infected with influenza A virus.
The team’s approach towards blood biomarker identification includes understanding the relationship between the peripheral blood and the initial site of infection (in this case, nasal region). In the NOVA challenge study, the team is identifying correlations between the activities of immune cells and viruses at these two sites. Similar to the previous studies, the cellular and viral data are analyzed alongside clinical symptom development from initial onset to end of the disease course.
Additionally, compared to the previous studies, the results from the Nova challenge study will be generalizable to a larger population. Study participants in this study included vaccinated and non-vaccinated individuals, as well as individuals with and without previous history of influenza.
(1) 2019, mBioIdentified three peripheral blood leukocyte gene expression phenotypes associated with active viral shedding, predicted length of shedding, or disease severity.
(a) Impact: These findings could contribute to the development of blood biomarkers
that can be used to predict
(b) Provide link to download pdf
(c) Provide link to learn more (Jessica will write feature that summarizes this work)
(2) 2021, Science Translational Medicine manipulates the immune system, causes widespread thrombosis that does not resolve, and targets signaling pathways that promote lung failure, fibrosis and impair tissue repair
(a) Provide link to download pdf
(b) Learn more at https://www.ncsa.illinois.edu/ncsa-researchers-co-author-article-on-the-damaging-effects-of-covid-19-on-lung-tissue/
Charles Blatti, PhD, Visual Analytics Group
[Contributions] Gene regulatory networks, machine learning, biological data visualization
Colleen Bushell, MFA, Visual Analytics Group, Healthcare Innovation Program Office
[Contributions] Information design, visual analytics, interactive workflows
John Kash, PhD, Viral Pathogenesis an Evolution Section at the NIH National Institute of Allergy andInfectious Diseases
[Contributions] biochemistry, cellular responses and mechanisms
Kathie Walters, PhD, Hood-Price Lab at the Institute for Systems Biology
[Contributions] virology, systems biology approaches
Ruoqing Zhui, PhD, Illinois Department of Statistics, NCSA
[Contributions] Dynamic treatment regimes, personalized medicine, machine learning