Density-based clustering analyses to identify heterogeneous cellular sub-populations

Tiffany M. Heaster, Alex J. Walsh, Bennett A. Landman, Melissa C. Skala

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

Autofluorescence microscopy of NAD(P)H and FAD provides functional metabolic measurements at the single-cell level. Here, density-based clustering algorithms were applied to metabolic autofluorescence measurements to identify cell-level heterogeneity in tumor cell cultures. The performance of the density-based clustering algorithm, DENCLUE, was tested in samples with known heterogeneity (co-cultures of breast carcinoma lines). DENCLUE was found to better represent the distribution of cell clusters compared to Gaussian mixture modeling. Overall, DENCLUE is a promising approach to quantify cell-level heterogeneity, and could be used to understand single cell population dynamics in cancer progression and treatment.

Original languageEnglish (US)
Title of host publicationDiagnosis and Treatment of Diseases in the Breast and Reproductive System
EditorsPaul J. Campagnola, Melissa C. Skala
PublisherSPIE
ISBN (Electronic)9781510605275
DOIs
StatePublished - 2017
EventDiagnosis and Treatment of Diseases in the Breast and Reproductive System - San Francisco, United States
Duration: Jan 28 2017Jan 29 2017

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume10043
ISSN (Print)1605-7422

Conference

ConferenceDiagnosis and Treatment of Diseases in the Breast and Reproductive System
CountryUnited States
CitySan Francisco
Period1/28/171/29/17

Keywords

  • breast cancer
  • cellular heterogeneity
  • density-based clustering
  • fluorescence lifetime
  • Metabolic imaging
  • quantitative spatial analysis

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Atomic and Molecular Physics, and Optics
  • Radiology Nuclear Medicine and imaging

Fingerprint Dive into the research topics of 'Density-based clustering analyses to identify heterogeneous cellular sub-populations'. Together they form a unique fingerprint.

Cite this