Assessing the utility of SELDI-TOF and model averaging for serum proteomic biomarker discovery

Sharad B Purohit, Robert Podolsky, Desmond Schatz, Andy Muir, Diane Hopkins, Yi Hua Huang, Jin-Xiong She

Research output: Contribution to journalArticle

30 Citations (Scopus)

Abstract

The SELDI-TOF technique was used to profile serum proteins from Type 1 diabetes (T1D) patients and healthy autoantibody-negative (AbN) controls. Univariate and multivariate analyses were performed to identify putative biomarkers for T1D and to assess the reproducibility of the SELDI technique. We found 146 protein/peptide peaks (581 total peaks discovered) in human serum showing statistical differences in expression levels between T1D patients and controls, with 84% of these peaks showing technical replication. Because individual proteins did not offer great power for disease prediction, we used our model averaging approach that combines the information from multiple multivariate models to accurately classify T1D and control subjects (88.9% specificity and 90.0% sensitivity). Analyses of a test subset of the data showed less accuracy (82.8% specificity and 76.2% sensitivity), although the results are still positive. Unfortunately, no multivariate model could be replicated using the same samples. This first attempt of high throughput analyses of the human serum proteome in T1D patients suggests that model averaging is a viable method for developing biomarkers; however, the reproducibility of SELDI-TOF is currently not sufficient to be used for classification of complex diseases like T1D.

Original languageEnglish (US)
Pages (from-to)6405-6415
Number of pages11
JournalProteomics
Volume6
Issue number24
DOIs
StatePublished - Dec 1 2006

Fingerprint

Biomarkers
Medical problems
Type 1 Diabetes Mellitus
Proteomics
Serum
Sensitivity and Specificity
Proteome
Autoantibodies
Blood Proteins
Proteins
Multivariate Analysis
Throughput
Peptides

Keywords

  • Biomarkers
  • Model
  • Prediction
  • Serum proteins
  • Type-1 Diabetes

ASJC Scopus subject areas

  • Biochemistry
  • Molecular Biology

Cite this

Assessing the utility of SELDI-TOF and model averaging for serum proteomic biomarker discovery. / Purohit, Sharad B; Podolsky, Robert; Schatz, Desmond; Muir, Andy; Hopkins, Diane; Huang, Yi Hua; She, Jin-Xiong.

In: Proteomics, Vol. 6, No. 24, 01.12.2006, p. 6405-6415.

Research output: Contribution to journalArticle

Purohit, Sharad B ; Podolsky, Robert ; Schatz, Desmond ; Muir, Andy ; Hopkins, Diane ; Huang, Yi Hua ; She, Jin-Xiong. / Assessing the utility of SELDI-TOF and model averaging for serum proteomic biomarker discovery. In: Proteomics. 2006 ; Vol. 6, No. 24. pp. 6405-6415.
@article{3f86e6a68f8a411eb698c23a77baeafc,
title = "Assessing the utility of SELDI-TOF and model averaging for serum proteomic biomarker discovery",
abstract = "The SELDI-TOF technique was used to profile serum proteins from Type 1 diabetes (T1D) patients and healthy autoantibody-negative (AbN) controls. Univariate and multivariate analyses were performed to identify putative biomarkers for T1D and to assess the reproducibility of the SELDI technique. We found 146 protein/peptide peaks (581 total peaks discovered) in human serum showing statistical differences in expression levels between T1D patients and controls, with 84{\%} of these peaks showing technical replication. Because individual proteins did not offer great power for disease prediction, we used our model averaging approach that combines the information from multiple multivariate models to accurately classify T1D and control subjects (88.9{\%} specificity and 90.0{\%} sensitivity). Analyses of a test subset of the data showed less accuracy (82.8{\%} specificity and 76.2{\%} sensitivity), although the results are still positive. Unfortunately, no multivariate model could be replicated using the same samples. This first attempt of high throughput analyses of the human serum proteome in T1D patients suggests that model averaging is a viable method for developing biomarkers; however, the reproducibility of SELDI-TOF is currently not sufficient to be used for classification of complex diseases like T1D.",
keywords = "Biomarkers, Model, Prediction, Serum proteins, Type-1 Diabetes",
author = "Purohit, {Sharad B} and Robert Podolsky and Desmond Schatz and Andy Muir and Diane Hopkins and Huang, {Yi Hua} and Jin-Xiong She",
year = "2006",
month = "12",
day = "1",
doi = "10.1002/pmic.200600420",
language = "English (US)",
volume = "6",
pages = "6405--6415",
journal = "Proteomics",
issn = "1615-9853",
publisher = "Wiley-VCH Verlag",
number = "24",

}

TY - JOUR

T1 - Assessing the utility of SELDI-TOF and model averaging for serum proteomic biomarker discovery

AU - Purohit, Sharad B

AU - Podolsky, Robert

AU - Schatz, Desmond

AU - Muir, Andy

AU - Hopkins, Diane

AU - Huang, Yi Hua

AU - She, Jin-Xiong

PY - 2006/12/1

Y1 - 2006/12/1

N2 - The SELDI-TOF technique was used to profile serum proteins from Type 1 diabetes (T1D) patients and healthy autoantibody-negative (AbN) controls. Univariate and multivariate analyses were performed to identify putative biomarkers for T1D and to assess the reproducibility of the SELDI technique. We found 146 protein/peptide peaks (581 total peaks discovered) in human serum showing statistical differences in expression levels between T1D patients and controls, with 84% of these peaks showing technical replication. Because individual proteins did not offer great power for disease prediction, we used our model averaging approach that combines the information from multiple multivariate models to accurately classify T1D and control subjects (88.9% specificity and 90.0% sensitivity). Analyses of a test subset of the data showed less accuracy (82.8% specificity and 76.2% sensitivity), although the results are still positive. Unfortunately, no multivariate model could be replicated using the same samples. This first attempt of high throughput analyses of the human serum proteome in T1D patients suggests that model averaging is a viable method for developing biomarkers; however, the reproducibility of SELDI-TOF is currently not sufficient to be used for classification of complex diseases like T1D.

AB - The SELDI-TOF technique was used to profile serum proteins from Type 1 diabetes (T1D) patients and healthy autoantibody-negative (AbN) controls. Univariate and multivariate analyses were performed to identify putative biomarkers for T1D and to assess the reproducibility of the SELDI technique. We found 146 protein/peptide peaks (581 total peaks discovered) in human serum showing statistical differences in expression levels between T1D patients and controls, with 84% of these peaks showing technical replication. Because individual proteins did not offer great power for disease prediction, we used our model averaging approach that combines the information from multiple multivariate models to accurately classify T1D and control subjects (88.9% specificity and 90.0% sensitivity). Analyses of a test subset of the data showed less accuracy (82.8% specificity and 76.2% sensitivity), although the results are still positive. Unfortunately, no multivariate model could be replicated using the same samples. This first attempt of high throughput analyses of the human serum proteome in T1D patients suggests that model averaging is a viable method for developing biomarkers; however, the reproducibility of SELDI-TOF is currently not sufficient to be used for classification of complex diseases like T1D.

KW - Biomarkers

KW - Model

KW - Prediction

KW - Serum proteins

KW - Type-1 Diabetes

UR - http://www.scopus.com/inward/record.url?scp=33846030136&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33846030136&partnerID=8YFLogxK

U2 - 10.1002/pmic.200600420

DO - 10.1002/pmic.200600420

M3 - Article

VL - 6

SP - 6405

EP - 6415

JO - Proteomics

JF - Proteomics

SN - 1615-9853

IS - 24

ER -