Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Feature-guided clustering of multi-dimensional flow cytometry datasets.

Qing T Zeng1, Juan Pablo Pratt, Jane Pak

  • 1Decision Systems Group, Brigham and Women's Hospital, 310 Thorn Building, 75 Francis Street, Harvard Medical School, Boston, MA 02115, USA. qzeng@dsg.harvard.edu

Journal of Biomedical Informatics
|August 12, 2006
PubMed
Summary

This study introduces a feature-guided (FG) algorithm for simplifying flow cytometry data. The FG method accurately clusters cells, outperforming traditional partition index methods in identifying cell populations.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Tumor-induced radial vascular patterning in the chick chorioallantoic membrane.

Microvascular research·2026
Same author

Intussusceptive angiogenesis: bridging in vivo and in vitro observations.

Angiogenesis·2025
Same author

Enhanced safety and potency in local angiogenic growth factor delivery via aptamer functionalization of hydrogels.

Journal of controlled release : official journal of the Controlled Release Society·2025
Same author

Machine learning-based risk scores are associated with conversion to dementia in Veterans.

Journal of Alzheimer's disease : JAD·2025
Same author

Identifying Probable Dementia in Undiagnosed Black and White Americans Using Machine Learning in Veterans Health Administration Electronic Health Records.

Big data and cognitive computing·2025
Same author

CLO25-075: Characterization of Lung Cancer Shedding and Spread Through Air Spaces in Lung Adenocarcinoma.

Journal of the National Comprehensive Cancer Network : JNCCN·2025

Area of Science:

  • Computational biology
  • Biotechnology
  • Data science

Background:

  • Flow cytometry generates complex, high-dimensional datasets detailing individual cell characteristics.
  • Simplifying these datasets by grouping similar cells into clusters is crucial for analysis.

Purpose of the Study:

  • To develop and validate a novel algorithm for clustering flow cytometry data.
  • To improve the accuracy and efficiency of cell population identification in complex datasets.

Main Methods:

  • A feature-guided (FG) k-means clustering algorithm was developed, driven by histogram features without pre-specifying the number of clusters.
  • The algorithm was tested on simulated cell-derived datasets using protein-coated microspheres with known populations.
  • Performance was compared against a partition index (PI) based cluster validity measure.

Related Experiment Videos

Main Results:

  • The FG approach successfully identified 100% of predetermined cell clusters in simulated data.
  • The PI method identified only 83.2% of clusters.
  • The FG method demonstrated significantly higher accuracy than PI in determining both the number of clusters and cell counts within clusters (p<.0001).

Conclusions:

  • Parameter feature analysis effectively guides k-means clustering for flow cytometry data.
  • The FG algorithm offers a more accurate and robust method for analyzing complex cytometry datasets.