A latent class model with hidden markov dependence for array CGH data

Stacia M. Desantis, E. Andrés Houseman, Brent A. Coull, David N. Louis, Gayatry Mohapatra, Rebecca Betensky

Research output: Contribution to journalArticle

Abstract

Array CGH is a high-throughput technique designed to detect genomic alterations linked to the development and progression of cancer. The technique yields fluorescence ratios that characterize DNA copy number change in tumor versus healthy cells. Classification of tumors based on aCGH profiles is of scientific interest but the analysis of these data is complicated by the large number of highly correlated measures. In this article, we develop a supervised Bayesian latent class approach for classification that relies on a hidden Markov model to account for the dependence in the intensity ratios. Supervision means that classification is guided by a clinical endpoint. Posterior inferences are made about class-specific copy number gains and losses. We demonstrate our technique on a study of brain tumors, for which our approach is capable of identifying subsets of tumors with different genomic profiles, and differentiates classes by survival much better than unsupervised methods.

Original languageEnglish (US)
Pages (from-to)1296-1305
Number of pages10
JournalBiometrics
Volume65
Issue number4
DOIs
StatePublished - Dec 1 2009

Fingerprint

Latent Class Model
Tumors
Tumor
neoplasms
Genomics
Brain Tumor
Latent Class
Neoplasms
DNA Copy Number Variations
Differentiate
Progression
Fluorescence
Markov Model
High Throughput
Cancer
genomics
Hidden Markov models
Brain Neoplasms
Brain
endpoints

Keywords

  • Array CGH
  • Hidden Markov Model
  • Latent class

ASJC Scopus subject areas

  • Applied Mathematics
  • Statistics and Probability
  • Agricultural and Biological Sciences(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Immunology and Microbiology(all)
  • Medicine(all)

Cite this

Desantis, S. M., Houseman, E. A., Coull, B. A., Louis, D. N., Mohapatra, G., & Betensky, R. (2009). A latent class model with hidden markov dependence for array CGH data. Biometrics, 65(4), 1296-1305. https://doi.org/10.1111/j.1541-0420.2009.01226.x

A latent class model with hidden markov dependence for array CGH data. / Desantis, Stacia M.; Houseman, E. Andrés; Coull, Brent A.; Louis, David N.; Mohapatra, Gayatry; Betensky, Rebecca.

In: Biometrics, Vol. 65, No. 4, 01.12.2009, p. 1296-1305.

Research output: Contribution to journalArticle

Desantis, SM, Houseman, EA, Coull, BA, Louis, DN, Mohapatra, G & Betensky, R 2009, 'A latent class model with hidden markov dependence for array CGH data', Biometrics, vol. 65, no. 4, pp. 1296-1305. https://doi.org/10.1111/j.1541-0420.2009.01226.x
Desantis SM, Houseman EA, Coull BA, Louis DN, Mohapatra G, Betensky R. A latent class model with hidden markov dependence for array CGH data. Biometrics. 2009 Dec 1;65(4):1296-1305. https://doi.org/10.1111/j.1541-0420.2009.01226.x
Desantis, Stacia M. ; Houseman, E. Andrés ; Coull, Brent A. ; Louis, David N. ; Mohapatra, Gayatry ; Betensky, Rebecca. / A latent class model with hidden markov dependence for array CGH data. In: Biometrics. 2009 ; Vol. 65, No. 4. pp. 1296-1305.
@article{162dac22b6c0491ebdc6988bffa994c0,
title = "A latent class model with hidden markov dependence for array CGH data",
abstract = "Array CGH is a high-throughput technique designed to detect genomic alterations linked to the development and progression of cancer. The technique yields fluorescence ratios that characterize DNA copy number change in tumor versus healthy cells. Classification of tumors based on aCGH profiles is of scientific interest but the analysis of these data is complicated by the large number of highly correlated measures. In this article, we develop a supervised Bayesian latent class approach for classification that relies on a hidden Markov model to account for the dependence in the intensity ratios. Supervision means that classification is guided by a clinical endpoint. Posterior inferences are made about class-specific copy number gains and losses. We demonstrate our technique on a study of brain tumors, for which our approach is capable of identifying subsets of tumors with different genomic profiles, and differentiates classes by survival much better than unsupervised methods.",
keywords = "Array CGH, Hidden Markov Model, Latent class",
author = "Desantis, {Stacia M.} and Houseman, {E. Andr{\'e}s} and Coull, {Brent A.} and Louis, {David N.} and Gayatry Mohapatra and Rebecca Betensky",
year = "2009",
month = "12",
day = "1",
doi = "10.1111/j.1541-0420.2009.01226.x",
language = "English (US)",
volume = "65",
pages = "1296--1305",
journal = "Biometrics",
issn = "0006-341X",
publisher = "Wiley-Blackwell",
number = "4",

}

TY - JOUR

T1 - A latent class model with hidden markov dependence for array CGH data

AU - Desantis, Stacia M.

AU - Houseman, E. Andrés

AU - Coull, Brent A.

AU - Louis, David N.

AU - Mohapatra, Gayatry

AU - Betensky, Rebecca

PY - 2009/12/1

Y1 - 2009/12/1

N2 - Array CGH is a high-throughput technique designed to detect genomic alterations linked to the development and progression of cancer. The technique yields fluorescence ratios that characterize DNA copy number change in tumor versus healthy cells. Classification of tumors based on aCGH profiles is of scientific interest but the analysis of these data is complicated by the large number of highly correlated measures. In this article, we develop a supervised Bayesian latent class approach for classification that relies on a hidden Markov model to account for the dependence in the intensity ratios. Supervision means that classification is guided by a clinical endpoint. Posterior inferences are made about class-specific copy number gains and losses. We demonstrate our technique on a study of brain tumors, for which our approach is capable of identifying subsets of tumors with different genomic profiles, and differentiates classes by survival much better than unsupervised methods.

AB - Array CGH is a high-throughput technique designed to detect genomic alterations linked to the development and progression of cancer. The technique yields fluorescence ratios that characterize DNA copy number change in tumor versus healthy cells. Classification of tumors based on aCGH profiles is of scientific interest but the analysis of these data is complicated by the large number of highly correlated measures. In this article, we develop a supervised Bayesian latent class approach for classification that relies on a hidden Markov model to account for the dependence in the intensity ratios. Supervision means that classification is guided by a clinical endpoint. Posterior inferences are made about class-specific copy number gains and losses. We demonstrate our technique on a study of brain tumors, for which our approach is capable of identifying subsets of tumors with different genomic profiles, and differentiates classes by survival much better than unsupervised methods.

KW - Array CGH

KW - Hidden Markov Model

KW - Latent class

UR - http://www.scopus.com/inward/record.url?scp=70450242678&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70450242678&partnerID=8YFLogxK

U2 - 10.1111/j.1541-0420.2009.01226.x

DO - 10.1111/j.1541-0420.2009.01226.x

M3 - Article

VL - 65

SP - 1296

EP - 1305

JO - Biometrics

JF - Biometrics

SN - 0006-341X

IS - 4

ER -