Clinical and Single-Cell Transcriptomics for Pneumonia Codeathon
Hosted by NIAID, CZI, BV-BRC, and Northwestern University
Chicago, IL – September 25-28, 2023
Apply Today: https://bit.ly/nu-codeathon
Introduction: This codeathon aims to address challenges associated with clinical data integration and modeling in the context of severe pneumonia. The concept originated within the Modeling Working Group established by the Systems Biology Consortium for Infectious Diseases supported by the National Institute of Allergy and Infectious Diseases (NIAID). The event is hosted by Northwestern University’s Successful Clinical Response In Pneumonia Therapy Systems Biology Center (SCRIPT), and offers the opportunity to collaborate with national and international scholars. The overall goal is to integrate clinical and single-cell transcriptomic data in ways that will improve patient outcome predictions and treatment responses for mechanically ventilated patients with suspected pneumonia.
Details: The in-person codeathon will take place in Chicago from September 25-28, concluding with a presentation of prototype software and/or analysis results during the NIAID Systems Biology Centers’ annual meeting on September 28. Participants can choose to develop their prototype or initial analysis into a software package or research publication after the codeathon.
Data: Participants will have access to 281 unpublished neutrophil-depleted single-cell transcriptomic data sets produced using the 10X Genomics Chromium platform from bronchoalveolar lavage samples of 183 severely ill pneumonia and non-pneumonia patients (average 5,500 UMIs/cell; 28 major cell types), along with 66 clinical measurements and demographic information (Gao et al. JCI, 2023, PhysioNet). Among these patients, 58 have transcriptomic samples from multiple days.
Additional Data Resources: Participants may choose to supplement study data with other open-source and accessible resources, including the CZ CellxGene repository that contains hundreds of standardized data collections from millions of cells characterizing the functionality of healthy and diseased human tissues, accessible via cellxgene-census API.
Projects: Groups will tackle challenges in clinical and transcriptomic data integration, such as mortality prediction, clinical state prediction, bacterial superinfection in viral pneumonia, analytical approach comparison, and public data resource integration. This codeathon affords a unique opportunity to analyze single cell transcriptomic responses in a complex environment with an acute illness.
Who Can Participate: Participants should have programming knowledge and prior data exposure. Experience with biological or clinical data is a plus. Our audience includes data scientists, clinical researchers, bioinformaticians, and systems biologists. Data use agreements will be required from participant’s institutions upon acceptance.
Registration Process: Selection priority will be given to participants who submit applications (https://bit.ly/nu-codeathon) by July 17. Participation is free, and virtual computing environments will be provided onsite at Northwestern University’s Chicago campus. Additional funds are available to cover travel expenses to the event. Applicants will be informed about the acceptance decision by August 4.
Selection: Participants will be selected based on motivation, ability to work with data and/or coding, diversity, and benefits to career of applicants. An independent steering committee will select participants and form five groups, each comprised of five participants, based on individual interests, backgrounds, and relevant expertise to work on each project.
Outcomes: All participants will collaborate with peers from other institutions and receive co-authorship on resulting manuscripts. After the codeathon, participants who write the main draft and prepare figures will be eligible for first authorship, while those contributing to the manuscript will be eligible for co-second authorship. Awards may be given.
Speakers and Mentors: SCRIPT’s clinical and data science staff will provide assistance during the codeathon, with preparations made ahead of the event. Afterward, Northwestern faculty will guide the completion of codeathon projects into research manuscripts or software packages. Speakers on September 28 include Kristian Andersen, PhD, Adolfo Garcia-Sastre, PhD, and Richard Wunderink, MD.
Partners and Sponsors: The National Institutes of Allergy and Infectious Diseases (NIAID) Systems Biology for Infectious Diseases Program, the Chan Zuckerberg Initiative (CZI), the Bacterial and Viral Bioinformatics Resource Center (BV-BRC), the Simpson Querrey Lung Institute for Translational Science (SQLIFTS) at Northwestern University Feinberg School of Medicine and the Canning Thoracic Institute at Northwestern Medicine.
Conclusion: The ability to leverage unique datasets presents an opportunity to address challenges associated with inferring clinical outcomes from single cell transcriptomic data and corresponding metadata. It also offers the potential for future collaborations in a unique environment. If you are interested in participating, pitching an idea for a team project, and/or serving as a team leader, we encourage you to apply. For queries, contact katie.clepp@northwestern.edu.
