Limits...
An integrated quantification method to increase the precision, robustness, and resolution of protein measurement in human plasma samples.

Li XJ, Lee LW, Hayward C, Brusniak MY, Fong PY, McLean M, Mulligan J, Spicer D, Fang KC, Hunsucker SW, Kearney P - Clin Proteomics (2015)

Bottom Line: Nevertheless, plasma proteins in low ng/ml to low μg/ml concentrations were measured with a median technical coefficient of variation (CV) of 11.9% using InteQuan.The corresponding median CV using SISQuan was 15.3% after linear fitting.We demonstrated that InteQuan is a simple yet robust quantification method for MS-based quantitative proteomics, especially for applications in biomarker research and in routine clinical testing.

View Article: PubMed Central - PubMed

Affiliation: Integrated Diagnostics, 219 Terry Avenue North, Suite 100, 98109 Seattle, WA USA.

ABSTRACT

Background: Current quantification methods for mass spectrometry (MS)-based proteomics either do not provide sufficient control of variability or are difficult to implement for routine clinical testing.

Results: We present here an integrated quantification (InteQuan) method that better controls pre-analytical and analytical variability than the popular quantification method using stable isotope-labeled standard peptides (SISQuan). We quantified 16 lung cancer biomarker candidates in human plasma samples in three assessment studies, using immunoaffinity depletion coupled with multiple reaction monitoring (MRM) MS. InteQuan outperformed SISQuan in precision in all three studies and tolerated a two-fold difference in sample loading. The three studies lasted over six months and encountered major changes in experimental settings. Nevertheless, plasma proteins in low ng/ml to low μg/ml concentrations were measured with a median technical coefficient of variation (CV) of 11.9% using InteQuan. The corresponding median CV using SISQuan was 15.3% after linear fitting. Furthermore, InteQuan surpassed SISQuan in measuring biological difference among clinical samples and in distinguishing benign versus cancer plasma samples.

Conclusions: We demonstrated that InteQuan is a simple yet robust quantification method for MS-based quantitative proteomics, especially for applications in biomarker research and in routine clinical testing.

No MeSH data available.


Related in: MedlinePlus

Calculation of generalized coefficient of variation (CV). (A-D) Results of all six clinical samples in Study III. (E-H) Results of the 29 HPS samples across all three studies. (A, E) Average InteQuan abundance versus experimental InteQuan abundance of individual proteins in individual samples. (B, F) Fitted SISQuan abundance versus experimental SISQuan abundance of individual proteins in individual samples. (C, G) The standard CV versus the generalized CV of InteQuan abundance. (D, H) The standard CV versus the generalized CV of SISQuan abundance.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4363461&req=5

Fig4: Calculation of generalized coefficient of variation (CV). (A-D) Results of all six clinical samples in Study III. (E-H) Results of the 29 HPS samples across all three studies. (A, E) Average InteQuan abundance versus experimental InteQuan abundance of individual proteins in individual samples. (B, F) Fitted SISQuan abundance versus experimental SISQuan abundance of individual proteins in individual samples. (C, G) The standard CV versus the generalized CV of InteQuan abundance. (D, H) The standard CV versus the generalized CV of SISQuan abundance.

Mentions: The high CVs of the target proteins using SISQuan in Study III reflected the large difference in the total protein concentration (Additional file5: Figure S3) rather than the precision of SISQuan. To compare the precision of InteQuan and SISQuan, a generalized method for CV calculation was developed. This method included two steps: First, the abundance of proteins in a sample was modeled either as linear functions of the loading volume (SISQuan) or as constants independent of the loading volume (InteQuan). Second, error propagation theory was applied to calculate the generalized CV as the standard deviation of differences between the modeled and the experimental abundances after logarithmic transformation. The modeled and the experimental abundances of all proteins in all samples collapsed nicely onto the respective diagonal line in Figure 4A and B, indicating that the method worked very well for both InteQuan and SISQuan. For SISQuan, it also demonstrated that proteins were measured within the respective linear dynamic range of the assays at all three concentrations. The generalized CVs and the standard CVs of InteQuan abundance were almost identical for all proteins in all samples (Figure 4C). On the contrary, the generalized CVs of SISQuan abundance were uniformly lower than the corresponding standard CVs (Figure 4D).Figure 4


An integrated quantification method to increase the precision, robustness, and resolution of protein measurement in human plasma samples.

Li XJ, Lee LW, Hayward C, Brusniak MY, Fong PY, McLean M, Mulligan J, Spicer D, Fang KC, Hunsucker SW, Kearney P - Clin Proteomics (2015)

Calculation of generalized coefficient of variation (CV). (A-D) Results of all six clinical samples in Study III. (E-H) Results of the 29 HPS samples across all three studies. (A, E) Average InteQuan abundance versus experimental InteQuan abundance of individual proteins in individual samples. (B, F) Fitted SISQuan abundance versus experimental SISQuan abundance of individual proteins in individual samples. (C, G) The standard CV versus the generalized CV of InteQuan abundance. (D, H) The standard CV versus the generalized CV of SISQuan abundance.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4363461&req=5

Fig4: Calculation of generalized coefficient of variation (CV). (A-D) Results of all six clinical samples in Study III. (E-H) Results of the 29 HPS samples across all three studies. (A, E) Average InteQuan abundance versus experimental InteQuan abundance of individual proteins in individual samples. (B, F) Fitted SISQuan abundance versus experimental SISQuan abundance of individual proteins in individual samples. (C, G) The standard CV versus the generalized CV of InteQuan abundance. (D, H) The standard CV versus the generalized CV of SISQuan abundance.
Mentions: The high CVs of the target proteins using SISQuan in Study III reflected the large difference in the total protein concentration (Additional file5: Figure S3) rather than the precision of SISQuan. To compare the precision of InteQuan and SISQuan, a generalized method for CV calculation was developed. This method included two steps: First, the abundance of proteins in a sample was modeled either as linear functions of the loading volume (SISQuan) or as constants independent of the loading volume (InteQuan). Second, error propagation theory was applied to calculate the generalized CV as the standard deviation of differences between the modeled and the experimental abundances after logarithmic transformation. The modeled and the experimental abundances of all proteins in all samples collapsed nicely onto the respective diagonal line in Figure 4A and B, indicating that the method worked very well for both InteQuan and SISQuan. For SISQuan, it also demonstrated that proteins were measured within the respective linear dynamic range of the assays at all three concentrations. The generalized CVs and the standard CVs of InteQuan abundance were almost identical for all proteins in all samples (Figure 4C). On the contrary, the generalized CVs of SISQuan abundance were uniformly lower than the corresponding standard CVs (Figure 4D).Figure 4

Bottom Line: Nevertheless, plasma proteins in low ng/ml to low μg/ml concentrations were measured with a median technical coefficient of variation (CV) of 11.9% using InteQuan.The corresponding median CV using SISQuan was 15.3% after linear fitting.We demonstrated that InteQuan is a simple yet robust quantification method for MS-based quantitative proteomics, especially for applications in biomarker research and in routine clinical testing.

View Article: PubMed Central - PubMed

Affiliation: Integrated Diagnostics, 219 Terry Avenue North, Suite 100, 98109 Seattle, WA USA.

ABSTRACT

Background: Current quantification methods for mass spectrometry (MS)-based proteomics either do not provide sufficient control of variability or are difficult to implement for routine clinical testing.

Results: We present here an integrated quantification (InteQuan) method that better controls pre-analytical and analytical variability than the popular quantification method using stable isotope-labeled standard peptides (SISQuan). We quantified 16 lung cancer biomarker candidates in human plasma samples in three assessment studies, using immunoaffinity depletion coupled with multiple reaction monitoring (MRM) MS. InteQuan outperformed SISQuan in precision in all three studies and tolerated a two-fold difference in sample loading. The three studies lasted over six months and encountered major changes in experimental settings. Nevertheless, plasma proteins in low ng/ml to low μg/ml concentrations were measured with a median technical coefficient of variation (CV) of 11.9% using InteQuan. The corresponding median CV using SISQuan was 15.3% after linear fitting. Furthermore, InteQuan surpassed SISQuan in measuring biological difference among clinical samples and in distinguishing benign versus cancer plasma samples.

Conclusions: We demonstrated that InteQuan is a simple yet robust quantification method for MS-based quantitative proteomics, especially for applications in biomarker research and in routine clinical testing.

No MeSH data available.


Related in: MedlinePlus