Pooling serum samples may lead to loss of potential biomarkers in SELDI-ToF MS proteomic profiling.

S Tariq Sadiq,Dan Agranoff

doi:10.1186/1477-5956-6-16

Abstract

BackgroundHigh throughput proteomic technology offers promise for the detection of disease biomarkers and proteomic signature patterns but biomarker discovery studies can be limited by cost factors when large sample size numbers are required. Pooling sera or plasma samples from disease cases potentially offers a solution to cost implications by reducing the standard errors of mass to charge values. Surface enhanced laser desorption/ionization time of flight (SELDI-ToF) mass spectra obtained from individual and pooled sera from invasive aspergillosis cases and controls were compared.ResultsPooling resulted in 50% loss of peak clusters detected in individual samples. Overall, loss was greatest for low intensity clusters. Peak intensities and case:control intensity ratios, among clusters not lost, demonstrated good reproducibility.ConclusionPooling sera results in significant potential biomarker loss when using SELDI-ToF MS.

Highlights

High throughput proteomic technology offers promise for the detection of disease biomarkers and proteomic signature patterns but biomarker discovery studies can be limited by cost factors when large sample size numbers are required
The aim of this study was to investigate the consistency of potential biomarker detection and variation in intensities for individual peak clusters obtained from SELDI-ToF before and after pooling serum samples
Where peaks in the pooled spectra were defined independently from peaks in the individual spectra, 197 and 110 peak clusters were detected from the individual and pooled sample experiments respectively. 97 (49.2%) peak clusters fulfilled the pre-defined criterion of being within 0.3% of their m/z values between pooled and individual samples and were retained for comparative analysis

Summary

Introduction

High throughput proteomic technology offers promise for the detection of disease biomarkers and proteomic signature patterns but biomarker discovery studies can be limited by cost factors when large sample size numbers are required. Pooling sera or plasma samples from disease cases potentially offers a solution to cost implications by reducing the standard errors of mass to charge values. Surface enhanced laser desorption/ionization time of flight (SELDI-ToF) mass spectra obtained from individual and pooled sera from invasive aspergillosis cases and controls were compared. In theory by pooling equal volumes of sera from cases and from controls, the potential loss of statistical power arising from reduced sample size is theoretically offset by the reduction in the standard error of the means of individual protein mass to charge (m/z) values (peak clusters), as long as the reduction in sample size is not too marked. It is unclear as to whether the process of pooling itself results in simple averaging of protein concentrations because of potential biological or bio-

Objectives

Methods

Results

Discussion

Conclusion