In an ‘Innovation’ article, published March 25, 2014 in Nature Reviews Immunology, Damien Chaussabel, PhD, (Benaroya Research Institute) and Nicole Baldwin, PhD, (Baylor Institute for Immunology Research) were invited to present the original data mining strategy they have developed for the analysis and interpretation of large-scale immunology datasets. Their modular repertoire analysis approach stands to significantly accelerate immune research advances through its ability to both simplify and fully leverage the large datasets being generated in systems immunology research.
This article comes on the heels of a research paper from Dr. Chaussabel’s laboratory that utilized the modular repertoire approach to investigate interferon signatures in the peripheral blood of lupus patients. The article was published last week on March 18, 2014 in Arthritis and Rheumatism alongside an editorial written by Peter Gregersen, MD, and Michaela Oswald, PhD, who are both research investigators at the Feinstein Institute in New York.
“Compared to the early days of peripheral gene expression analysis, this is impressive progress. This has been dependent on the creativity and commitment of (Chaussabel and Baldwin), but it is also important to emphasize the role of open data sharing in this process,” noted Drs. Gregersen and Oswald in the Arthritis and Rheumatism editorial. “Another aspect of this is the commitment of Chaussabel, and others to make data available in a format that is flexible and easy to understand by immunologists and biologists who do not have advanced computational skills.”
Over the past several years, the use of large-scale profiling assays in immunology has allowed researchers to obtain extensive molecular and cellular data associated with immune responses that play a critical role in both health maintenance and disease. However, as the amount of data being collected has increased via these technological breakthroughs, the analyses needed to mine the data have become increasingly complex and time consuming. Drs. Chaussabel and Baldwin’s modular transcriptional repertoire analysis approach is a significant achievement for systems-based research because not only does it simplify analysis of large-scale datasets but it also allows data from multiple studies across multiple diseases or groups to be combined and leveraged together. Integrated with this approach, web applications by the BRI software core led by Charlie Quinn, as well as Scott Presnell, PhD, and Elizabeth Whalen, PhD, both bioinformaticians from the BRI systems immunology division, allow users to further explore modular fingerprints, adjust parameters and access gene or subject level data and interpretations. Access to the web application is available via interactive figures in the article.
Additionally, while Drs. Chaussabel and Baldwin describe this approach and its application in immunology research, the methodology could translate beyond immunology into other biological fields wherein large amounts of data are being generated, such as infectious disease or cancer research. Indeed, the implementation of such data mining approaches pave the way for the development of comprehensive analytic solutions that will dramatically accelerate knowledge discovery from collective biomedical Big Data.
