The cell graphs of cancer

Cigdem Gunduz, Bülent Yener, Sakir Gultekin

Research output: Contribution to journalArticle

88 Citations (Scopus)

Abstract

We report a novel, proof-of-concept, computational method that models a type of brain cancer (glioma) only by using the topological properties of its cells in the tissue image. From low-magnification (80x) tissue images of 384 × 384 pixels, we construct the graphs of the cells based on the locations of the cells within the images. We generate such cell graphs of 1000-3000 cells (nodes) with 2000-10 000 links, each of which is calculated as a decaying exponential function of the Euclidean distance between every pair of cells in accordance with the Waxman model. At the cellular level, we compute the graph metrics of the cell graphs, including the degree, clustering coefficient, eccentricity and closeness for each cell. Working with a total of 285 tissue samples surgically removed from 12 different patients, we demonstrate that the self-organizing clusters of cancerous cells exhibit distinctive graph metrics that distinguish them from the healthy cells and the unhealthy inflamed cells at the cellular level with an accuracy of at least 85%. At the tissue level, we accomplish correct tissue classifications of cancerous, healthy and non-neoplastic inflamed tissue samples with an accuracy of 100% by requiring correct classification for the majority of the cells within the tissue sample.

Original languageEnglish (US)
JournalBioinformatics
Volume20
Issue numberSUPPL. 1
DOIs
StatePublished - 2004
Externally publishedYes

Fingerprint

Cancer
Tissue
Cell
Graph in graph theory
Neoplasms
Metric Graphs
Cells
Exponential functions
Computational methods
Brain
Pixels
Clustering Coefficient
Eccentricity
Self-organizing
Euclidean Distance
Topological Properties
Brain Neoplasms
Glioma
Computational Methods
Cluster Analysis

ASJC Scopus subject areas

  • Clinical Biochemistry
  • Computer Science Applications
  • Computational Theory and Mathematics

Cite this

Gunduz, C., Yener, B., & Gultekin, S. (2004). The cell graphs of cancer. Bioinformatics, 20(SUPPL. 1). https://doi.org/10.1093/bioinformatics/bth933

The cell graphs of cancer. / Gunduz, Cigdem; Yener, Bülent; Gultekin, Sakir.

In: Bioinformatics, Vol. 20, No. SUPPL. 1, 2004.

Research output: Contribution to journalArticle

Gunduz, C, Yener, B & Gultekin, S 2004, 'The cell graphs of cancer', Bioinformatics, vol. 20, no. SUPPL. 1. https://doi.org/10.1093/bioinformatics/bth933
Gunduz, Cigdem ; Yener, Bülent ; Gultekin, Sakir. / The cell graphs of cancer. In: Bioinformatics. 2004 ; Vol. 20, No. SUPPL. 1.
@article{74bd524c22a84d508009a8c1b3899006,
title = "The cell graphs of cancer",
abstract = "We report a novel, proof-of-concept, computational method that models a type of brain cancer (glioma) only by using the topological properties of its cells in the tissue image. From low-magnification (80x) tissue images of 384 × 384 pixels, we construct the graphs of the cells based on the locations of the cells within the images. We generate such cell graphs of 1000-3000 cells (nodes) with 2000-10 000 links, each of which is calculated as a decaying exponential function of the Euclidean distance between every pair of cells in accordance with the Waxman model. At the cellular level, we compute the graph metrics of the cell graphs, including the degree, clustering coefficient, eccentricity and closeness for each cell. Working with a total of 285 tissue samples surgically removed from 12 different patients, we demonstrate that the self-organizing clusters of cancerous cells exhibit distinctive graph metrics that distinguish them from the healthy cells and the unhealthy inflamed cells at the cellular level with an accuracy of at least 85{\%}. At the tissue level, we accomplish correct tissue classifications of cancerous, healthy and non-neoplastic inflamed tissue samples with an accuracy of 100{\%} by requiring correct classification for the majority of the cells within the tissue sample.",
author = "Cigdem Gunduz and B{\"u}lent Yener and Sakir Gultekin",
year = "2004",
doi = "10.1093/bioinformatics/bth933",
language = "English (US)",
volume = "20",
journal = "Bioinformatics",
issn = "1367-4803",
publisher = "Oxford University Press",
number = "SUPPL. 1",

}

TY - JOUR

T1 - The cell graphs of cancer

AU - Gunduz, Cigdem

AU - Yener, Bülent

AU - Gultekin, Sakir

PY - 2004

Y1 - 2004

N2 - We report a novel, proof-of-concept, computational method that models a type of brain cancer (glioma) only by using the topological properties of its cells in the tissue image. From low-magnification (80x) tissue images of 384 × 384 pixels, we construct the graphs of the cells based on the locations of the cells within the images. We generate such cell graphs of 1000-3000 cells (nodes) with 2000-10 000 links, each of which is calculated as a decaying exponential function of the Euclidean distance between every pair of cells in accordance with the Waxman model. At the cellular level, we compute the graph metrics of the cell graphs, including the degree, clustering coefficient, eccentricity and closeness for each cell. Working with a total of 285 tissue samples surgically removed from 12 different patients, we demonstrate that the self-organizing clusters of cancerous cells exhibit distinctive graph metrics that distinguish them from the healthy cells and the unhealthy inflamed cells at the cellular level with an accuracy of at least 85%. At the tissue level, we accomplish correct tissue classifications of cancerous, healthy and non-neoplastic inflamed tissue samples with an accuracy of 100% by requiring correct classification for the majority of the cells within the tissue sample.

AB - We report a novel, proof-of-concept, computational method that models a type of brain cancer (glioma) only by using the topological properties of its cells in the tissue image. From low-magnification (80x) tissue images of 384 × 384 pixels, we construct the graphs of the cells based on the locations of the cells within the images. We generate such cell graphs of 1000-3000 cells (nodes) with 2000-10 000 links, each of which is calculated as a decaying exponential function of the Euclidean distance between every pair of cells in accordance with the Waxman model. At the cellular level, we compute the graph metrics of the cell graphs, including the degree, clustering coefficient, eccentricity and closeness for each cell. Working with a total of 285 tissue samples surgically removed from 12 different patients, we demonstrate that the self-organizing clusters of cancerous cells exhibit distinctive graph metrics that distinguish them from the healthy cells and the unhealthy inflamed cells at the cellular level with an accuracy of at least 85%. At the tissue level, we accomplish correct tissue classifications of cancerous, healthy and non-neoplastic inflamed tissue samples with an accuracy of 100% by requiring correct classification for the majority of the cells within the tissue sample.

UR - http://www.scopus.com/inward/record.url?scp=24744456340&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=24744456340&partnerID=8YFLogxK

U2 - 10.1093/bioinformatics/bth933

DO - 10.1093/bioinformatics/bth933

M3 - Article

VL - 20

JO - Bioinformatics

JF - Bioinformatics

SN - 1367-4803

IS - SUPPL. 1

ER -