Immune monitoring

Oral

Dafi: A Novel and Robust Auto-gating Approach to Computational Identification of Cell Populations from High-dimensional Flow Cytometry Data

Friday, June 16
6:15 PM - 7:30 PM

Background: Manual gating analysis of cytometry data is subjective, irreproducible, and inaccurate. The dimensionality of cytometry data keeps increasing, making auto-gating methods both desirable and essential. However, the assessment and validation of the results from auto-gating methods is highly challenging, making the adoption of auto-gating approaches nontrivial.


Methods: This work presents a new computational approach – DAFi (Directed Automated Filtering and Identification of Cell Populations) for robust auto-gating analysis of high-dimensional flow cytometry data. DAFi is designed and implemented with multiple auto-gating options, including clustering, bisecting, slope-based, reversed-gating, and their combinations to identify different types of cell populations defined in user manual gating strategy. Through mapping high-dimensional data clusters into user-defined gating hierarchy, DAFi not only preserves the data clusters generated in unsupervised learning, but also makes the results easily interpretable as in supervised methods.


Results and Conclusions: Both quantitative assessment and visual examination of dot plots are used to evaluate the performance of DAFi against manual gating analysis across a variety of translational research projects as well as the diagnostics of blood cancers (AML and CLL). The results show that DAFi not only preserves natural shapes of both major and small lymphocyte subsets in an interpretable and reproducible way but also identifies novel cell subsets that differ between subject groups, which manual gating cannot find. Based on the results, we conclude that DAFi is highly promising to substitute manual gating for reproducible and exploratory analysis of high-dimensional flow cytometry data for biomarker identification in translational and clinical immunology research.

Yu Qian

Assistant Professor
J. Craig Venter Institute

Presentation(s):

    Send Email for Yu Qian

    Julie Burel

    La Jolla Institute for Allergy and Immunology

    Presentation(s):

      Send Email for Julie Burel

      Alexandra Lee

      J. Craig Venter Institute

      Presentation(s):

        Send Email for Alexandra Lee

        Ivan Chang

        J. Craig Venter Institute

        Presentation(s):

          Send Email for Ivan Chang

          Cecilia Lindestam Arlehamn

          La Jolla Institute for Allergy and Immunology

          Presentation(s):

            Send Email for Cecilia Lindestam Arlehamn

            Daniela Weiskopf

            La Jolla Institute for Allergy and Immunology

            Presentation(s):

              Send Email for Daniela Weiskopf

              Bjoern Peters

              Associate Professor
              La Jolla Institute for Allergy and Immunology

              Presentation(s):

                Send Email for Bjoern Peters

                Alessandro Sette

                Professor
                La Jolla Institute for Allergy and Immunology

                Presentation(s):

                  Send Email for Alessandro Sette

                  Rick Stanton

                  Human Longevity Inc.

                  Presentation(s):

                    Send Email for Rick Stanton

                    Richard H. Scheuermann

                    Professor
                    J. Craig Venter Institute

                    Presentation(s):

                      Send Email for Richard Scheuermann


                      Assets

                      Dafi: A Novel and Robust Auto-gating Approach to Computational Identification of Cell Populations from High-dimensional Flow Cytometry Data



                      Attendees who have favorited this

                      Send Email for Dafi: A Novel and Robust Auto-gating Approach to Computational Identification of Cell Populations from High-dimensional Flow Cytometry Data