Statistical identification of differentially labeled peptides from liquid chromatography tandem mass spectrometry

Abstract

LC-MS/MS with certain labeling techniques such as isotope-coded affinity tag (ICAT) enables quantitative analysis of paired protein samples. However, current identification and quantification of differentially expressed peptides (and proteins) are not reliable for large proteomics screening of complex biological samples. The number of replicates is often limited because of the high cost of experiments and the limited supply of samples. Traditionally, a simple fold change cutoff is used, which results in a high rate of false positives. Standard statistical methods such as the two-sample ttest are unreliable and severely underpowered due to high variability in LC-MS/MS data, especially when only a small number of replicates are available. Using an advanced error pooling technique, we propose a novel statistical method that can reliably identify differentially expressed proteins while maintaining a high sensitivity, particularly with a small number of replicates. The proposed method was applied both to an extensive simulation study and a proteomics comparison between microparticles (MPs) generated from platelet (platelet MPs) and MPs isolated from plasma (plasma MPs). In these studies, we show a significant improvement of our statistical analysis in the identification of proteins that are differentially expressed but not detected by other statistical methods. In particular, several important proteins -two peptides for b-globin and three peptides for von Willebrand Factor (vWF) -were identified with very small false discovery rates (FDRs) by our method, while none was significant when other conventional methods were used. These proteins have been reported with their important roles in microparticles in human blood cells: vWF is a platelet and endothelial cell product that binds to P-selectin, GP1b, and GP IIb/IIIa, and b-globin is one of the peptides of hemoglobin involved in the transportation of oxygen by red blood cells.

An Introduction to Proteomics Technologies for the Genomics Scientist

“…is quickly becoming a major focus of investigation in quantitative proteomics studies (e.g. see Refs [23,40,41,68]). It is often the case that sample fractionation and enrichment strategies are necessary to provide greater depth of coverage.…”

“…This is mainly due to the fact that the analysis is performed on the peptide products of the starting mixture en masse without any measurement of the intact proteins, such that isoforms remain unresolved and proteolytic products are masked by the intact species if both are present. The lower statistical power stems mainly from the fact that the mass spectrometer is sampling ions for tandem MS based on relative signal intensities that can vary from run to run [40,41], requiring many technical replicates for each independent sample to attain acceptable levels of statistical power.…”

“…Analysis of surrogate peptides also enables the identification of proteins that are otherwise refractory to 2D gels and MALDI-IMS due to issues such as solubility, hydrophobicity and extreme MW/pI of the intact proteins. However, as mentioned above, important post-translational modifications and processed forms would only be detected if the important peptides are captured or directly targeted with this technology (because information on the intact protein is not retained), and it also becomes increasingly challenging to perform complex experimental comparisons that are relatively straightforward by DIGE or MALDI-IMS [23,40,41]. Whether this becomes a strength or a limitation depends on the experimental goals; for a global, discovery-based experiment these intact protein attributes can go undetected, but in a targeted approach, especially using MRM-directed MS approaches [39], specific modifications can be quantified under conditions where they might go undetected in the 2D gel or MALDI approaches.…”

“…Statistical power depends heavily on the analytical variation of the instrumentation making the measurement, as well as the number of independent (biological) replicates. For example, the experimental noise of DIGE is extremely low owing to the internal standard experimental design, enabling statistically powered experiments with very few biological replicates [23], whereas LC/MS experiments have a relatively high degree of analytical variation with little to no internal standard methodologies being employed, often resulting in underpowered experiments unless sufficient biological and technical replicates are employed [40,41].…”

Computational and statistical methods for high-throughput analysis of post-translational modifications of proteins

“…It is now established that fold-change thresholds cannot be used for determining DRFs with high confidence [78]. Methods to estimate significant fold-changes from the ratio distribution rely on the assumptions that first only a tiny fraction of the proteins is regulated and second the variance of each protein is given by the distribution width, preventing a rigorous evaluation of statistical significance.…”