The optimal exponent base for emPAI is 6.5.
Bottom Line:
EmPAI was first proposed by Ishihama et al [1] who found that PAI is approximately proportional to the logarithm of absolute protein concentration.I define emPAI65 = 6.5(PAI)-1 and show that it performs significantly better than emPAI, while it is equally easy to compute.I conclude that emPAI65 ought to be used instead of the original emPAI for protein quantitation.
View Article:
PubMed Central - PubMed
Affiliation: Department of Biochemistry and Molecular Biology, Sealy Center for Molecular Medicine, University of Texas Medical Branch, Galveston, Texas, United States of America. askudlic@utmb.edu
ABSTRACT
Show MeSH
Exponentially Modified Protein Abundance Index (emPAI) is an established method of estimating protein abundances from peptide counts in a single LC-MS/MS experiment. EmPAI is defined as 10(PAI) minus one, where PAI (Protein Abundance Index) denotes the ratio of observed to observable peptides. EmPAI was first proposed by Ishihama et al [1] who found that PAI is approximately proportional to the logarithm of absolute protein concentration. I define emPAI65 = 6.5(PAI)-1 and show that it performs significantly better than emPAI, while it is equally easy to compute. The higher accuracy of emPAI65 is demonstrated by analyzing three data sets, including the one used in the original study [1]. I conclude that emPAI65 ought to be used instead of the original emPAI for protein quantitation. |
Related In:
Results -
Collection
getmorefigures.php?uid=PMC3293797&req=5
Mentions: Here, I compute the generalized emPAI for the same 46 proteins for a wide range of exponent bases, from a = 3 to a = 15 with a step of 0.01. For each base, I estimate the best scaling factor to convert the relative abundances inferred from GemPAI into absolute concentrations, and next I calculate the deviation factors for all proteins. The average deviation factor as a function of the base is shown in Fig. 2. These results show that the average deviation factor is the lowest (corresponding to the best estimate of protein abundance by the generalized emPAI) for a = 6.50. Throughout this paper, I will denote GemPAI(*, 6.5) as emPAI65. EmPAI65 can be directly computed from PAI, or from emPAI using the following relation:(1)To independently demonstrate the superiority of emPAI65 over emPAI, I have computed the values of emPAI65 for the proteins identified in the large-scale proteome profiling experiment of [7], and related them to the protein concentrations in E. coli cells measured by [8], using 42 data points analogously to the comparison presented in Fig. 2 of [7]. This dataset has a very high dynamic range, with the measured protein abundances spanning four orders of magnitude. I have computed the deviation factors for both emPAI and emPAI65 for the proteins plotted in Fig. 2 of [7]. The average deviation factor is 4.72 for emPAI65 and 7.78 for emPAI, again significantly lower for quantitation using base 6.5 rather than base 10. The measured protein concentrations are plotted against estimates with emPAI and emPAI65 in Figure 3, showing the greater deviation from proportionality in case of the standard emPAI. Note that unlike the mouse lysate data of [1], the E. coli data are derived from experiments by two research groups and a larger variance is expected, which is reflected by a higher average deviation factor. For this reason I did not use this dataset in the initial determination of the optimal exponent base. |
View Article: PubMed Central - PubMed
Affiliation: Department of Biochemistry and Molecular Biology, Sealy Center for Molecular Medicine, University of Texas Medical Branch, Galveston, Texas, United States of America. askudlic@utmb.edu