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

Sadiq ST, Agranoff D - Proteome Sci (2008)

Bottom Line: 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.Pooling resulted in 50% loss of peak clusters detected in individual samples.Overall, loss was greatest for low intensity clusters.

View Article: PubMed Central - HTML - PubMed

Affiliation: Centre for Infection, St George's University of London, UK. ssadiq@sgul.ac.uk

ABSTRACT

Background: 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.

Results: Pooling 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.

Conclusion: Pooling sera results in significant potential biomarker loss when using SELDI-ToF MS.

No MeSH data available.


Related in: MedlinePlus

Representative SELDI spectra, illustrating a quality control spectrum, 'typical' individual case and control spectra and pooled sample spectra from the case and control groups. For each of the pooled sets, the 20 individual samples in each group were combined to form a pooled sample, from which a spectrum was generated under the same conditions as for individual samples.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2414484&req=5

Figure 2: Representative SELDI spectra, illustrating a quality control spectrum, 'typical' individual case and control spectra and pooled sample spectra from the case and control groups. For each of the pooled sets, the 20 individual samples in each group were combined to form a pooled sample, from which a spectrum was generated under the same conditions as for individual samples.

Mentions: Representative spectra obtained from quality control, individual case and control and pooled case and control spectra are shown in figure 2. Where peaks in the pooled spectra were defined independently from peaks in the individual spectra (set A), 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 (figure 3). Pooling resulted in the 'loss' of 100 peak clusters, while 13 peak clusters appeared to be unique to the pooled samples. In the quality control replicates, 44/55 (80%) of peaks with m/z > 2500 detected at a S/N ≥ 5 and a mass window of 0.3%, were present in all 7 replicates. Only a few clusters were generated across the mass range of 20,000 to 100,000 using a mass window of 0.3% and this hardly changed when using a mass window of 0.6% over the same range (data not shown).


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

Sadiq ST, Agranoff D - Proteome Sci (2008)

Representative SELDI spectra, illustrating a quality control spectrum, 'typical' individual case and control spectra and pooled sample spectra from the case and control groups. For each of the pooled sets, the 20 individual samples in each group were combined to form a pooled sample, from which a spectrum was generated under the same conditions as for individual samples.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2414484&req=5

Figure 2: Representative SELDI spectra, illustrating a quality control spectrum, 'typical' individual case and control spectra and pooled sample spectra from the case and control groups. For each of the pooled sets, the 20 individual samples in each group were combined to form a pooled sample, from which a spectrum was generated under the same conditions as for individual samples.
Mentions: Representative spectra obtained from quality control, individual case and control and pooled case and control spectra are shown in figure 2. Where peaks in the pooled spectra were defined independently from peaks in the individual spectra (set A), 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 (figure 3). Pooling resulted in the 'loss' of 100 peak clusters, while 13 peak clusters appeared to be unique to the pooled samples. In the quality control replicates, 44/55 (80%) of peaks with m/z > 2500 detected at a S/N ≥ 5 and a mass window of 0.3%, were present in all 7 replicates. Only a few clusters were generated across the mass range of 20,000 to 100,000 using a mass window of 0.3% and this hardly changed when using a mass window of 0.6% over the same range (data not shown).

Bottom Line: 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.Pooling resulted in 50% loss of peak clusters detected in individual samples.Overall, loss was greatest for low intensity clusters.

View Article: PubMed Central - HTML - PubMed

Affiliation: Centre for Infection, St George's University of London, UK. ssadiq@sgul.ac.uk

ABSTRACT

Background: 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.

Results: Pooling 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.

Conclusion: Pooling sera results in significant potential biomarker loss when using SELDI-ToF MS.

No MeSH data available.


Related in: MedlinePlus