Derivation of stable microarray cancer-differentiating signatures using consensus scoring of multiple random sampling and gene-ranking consistency evaluation

Qun Tang Zhi, Yi Han Lian, Huang Lin Hong, Juan Cui, Jia Jia, Chuan Low Boon, Wen Li Bao, Zong Chen Yu

Research output: Contribution to journalArticle

17 Citations (Scopus)

Abstract

Microarrays have been explored for deriving molecular signatures to determine disease outcomes, mechanisms, targets, and treatment strategies. Although exhibiting good predictive performance, some derived signatures are unstable due to noises arising from measurement variability and biological differences. Improvements in measurement, annotation, and signature selection methods have been proposed. We explored a new signature selection method that incorporates consensus scoring of multiple random sampling and multistep evaluation of gene-ranking consistency for maximally avoiding erroneous elimination of predictor genes. This method was tested by using a well-studied 62-sample colon cancer data set and two other cancer data sets (86-sample lung adenocarcinoma and 60-sample hepatocellular carcinoma). For the colon cancer data set, the derived signatures of 20 sampling sets, composed of 10,000 training test sets, are fairly stable with 80% of top 50 and 69% to 93% of all predictor genes shared by all 20 signatures. These shared predictor genes include 48 cancer-related and 16 cancer-implicated genes, as well as 50% of the previously derived predictor genes. The derived signatures outperform all previously derived signatures in predicting colon cancer outcomes from an independent data set collected from the Stanford Microarray Database. Our method showed similar performance for the other two data sets, suggesting its usefulness in deriving stable signatures for biomarker and target discovery.

Original languageEnglish (US)
Pages (from-to)9996-10003
Number of pages8
JournalCancer Research
Volume67
Issue number20
DOIs
StatePublished - Oct 15 2007

Fingerprint

Colonic Neoplasms
Genes
Neoplasms
Neoplasm Genes
Noise
Hepatocellular Carcinoma
Biomarkers
Datasets
Databases
Adenocarcinoma of lung

ASJC Scopus subject areas

  • Oncology
  • Cancer Research

Cite this

Derivation of stable microarray cancer-differentiating signatures using consensus scoring of multiple random sampling and gene-ranking consistency evaluation. / Zhi, Qun Tang; Lian, Yi Han; Hong, Huang Lin; Cui, Juan; Jia, Jia; Boon, Chuan Low; Bao, Wen Li; Yu, Zong Chen.

In: Cancer Research, Vol. 67, No. 20, 15.10.2007, p. 9996-10003.

Research output: Contribution to journalArticle

Zhi, Qun Tang ; Lian, Yi Han ; Hong, Huang Lin ; Cui, Juan ; Jia, Jia ; Boon, Chuan Low ; Bao, Wen Li ; Yu, Zong Chen. / Derivation of stable microarray cancer-differentiating signatures using consensus scoring of multiple random sampling and gene-ranking consistency evaluation. In: Cancer Research. 2007 ; Vol. 67, No. 20. pp. 9996-10003.
@article{f525a901d7044ad0b0b0d5cb2b14fa30,
title = "Derivation of stable microarray cancer-differentiating signatures using consensus scoring of multiple random sampling and gene-ranking consistency evaluation",
abstract = "Microarrays have been explored for deriving molecular signatures to determine disease outcomes, mechanisms, targets, and treatment strategies. Although exhibiting good predictive performance, some derived signatures are unstable due to noises arising from measurement variability and biological differences. Improvements in measurement, annotation, and signature selection methods have been proposed. We explored a new signature selection method that incorporates consensus scoring of multiple random sampling and multistep evaluation of gene-ranking consistency for maximally avoiding erroneous elimination of predictor genes. This method was tested by using a well-studied 62-sample colon cancer data set and two other cancer data sets (86-sample lung adenocarcinoma and 60-sample hepatocellular carcinoma). For the colon cancer data set, the derived signatures of 20 sampling sets, composed of 10,000 training test sets, are fairly stable with 80{\%} of top 50 and 69{\%} to 93{\%} of all predictor genes shared by all 20 signatures. These shared predictor genes include 48 cancer-related and 16 cancer-implicated genes, as well as 50{\%} of the previously derived predictor genes. The derived signatures outperform all previously derived signatures in predicting colon cancer outcomes from an independent data set collected from the Stanford Microarray Database. Our method showed similar performance for the other two data sets, suggesting its usefulness in deriving stable signatures for biomarker and target discovery.",
author = "Zhi, {Qun Tang} and Lian, {Yi Han} and Hong, {Huang Lin} and Juan Cui and Jia Jia and Boon, {Chuan Low} and Bao, {Wen Li} and Yu, {Zong Chen}",
year = "2007",
month = "10",
day = "15",
doi = "10.1158/0008-5472.CAN-07-1601",
language = "English (US)",
volume = "67",
pages = "9996--10003",
journal = "Cancer Research",
issn = "0008-5472",
publisher = "American Association for Cancer Research Inc.",
number = "20",

}

TY - JOUR

T1 - Derivation of stable microarray cancer-differentiating signatures using consensus scoring of multiple random sampling and gene-ranking consistency evaluation

AU - Zhi, Qun Tang

AU - Lian, Yi Han

AU - Hong, Huang Lin

AU - Cui, Juan

AU - Jia, Jia

AU - Boon, Chuan Low

AU - Bao, Wen Li

AU - Yu, Zong Chen

PY - 2007/10/15

Y1 - 2007/10/15

N2 - Microarrays have been explored for deriving molecular signatures to determine disease outcomes, mechanisms, targets, and treatment strategies. Although exhibiting good predictive performance, some derived signatures are unstable due to noises arising from measurement variability and biological differences. Improvements in measurement, annotation, and signature selection methods have been proposed. We explored a new signature selection method that incorporates consensus scoring of multiple random sampling and multistep evaluation of gene-ranking consistency for maximally avoiding erroneous elimination of predictor genes. This method was tested by using a well-studied 62-sample colon cancer data set and two other cancer data sets (86-sample lung adenocarcinoma and 60-sample hepatocellular carcinoma). For the colon cancer data set, the derived signatures of 20 sampling sets, composed of 10,000 training test sets, are fairly stable with 80% of top 50 and 69% to 93% of all predictor genes shared by all 20 signatures. These shared predictor genes include 48 cancer-related and 16 cancer-implicated genes, as well as 50% of the previously derived predictor genes. The derived signatures outperform all previously derived signatures in predicting colon cancer outcomes from an independent data set collected from the Stanford Microarray Database. Our method showed similar performance for the other two data sets, suggesting its usefulness in deriving stable signatures for biomarker and target discovery.

AB - Microarrays have been explored for deriving molecular signatures to determine disease outcomes, mechanisms, targets, and treatment strategies. Although exhibiting good predictive performance, some derived signatures are unstable due to noises arising from measurement variability and biological differences. Improvements in measurement, annotation, and signature selection methods have been proposed. We explored a new signature selection method that incorporates consensus scoring of multiple random sampling and multistep evaluation of gene-ranking consistency for maximally avoiding erroneous elimination of predictor genes. This method was tested by using a well-studied 62-sample colon cancer data set and two other cancer data sets (86-sample lung adenocarcinoma and 60-sample hepatocellular carcinoma). For the colon cancer data set, the derived signatures of 20 sampling sets, composed of 10,000 training test sets, are fairly stable with 80% of top 50 and 69% to 93% of all predictor genes shared by all 20 signatures. These shared predictor genes include 48 cancer-related and 16 cancer-implicated genes, as well as 50% of the previously derived predictor genes. The derived signatures outperform all previously derived signatures in predicting colon cancer outcomes from an independent data set collected from the Stanford Microarray Database. Our method showed similar performance for the other two data sets, suggesting its usefulness in deriving stable signatures for biomarker and target discovery.

UR - http://www.scopus.com/inward/record.url?scp=35448932757&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=35448932757&partnerID=8YFLogxK

U2 - 10.1158/0008-5472.CAN-07-1601

DO - 10.1158/0008-5472.CAN-07-1601

M3 - Article

C2 - 17942933

AN - SCOPUS:35448932757

VL - 67

SP - 9996

EP - 10003

JO - Cancer Research

JF - Cancer Research

SN - 0008-5472

IS - 20

ER -