|Home | About | Journals | Submit | Contact Us | Français|
Differential proteome studies are a powerful tool for the analysis of differences between two sample states. A challenge encountered in any proteome study is the reproducibility of the sample preparation and data analysis. The significance analysis of the results and the extent to which changes can reliably be detected are affected by this.
We studied the changes of the proteome during cell differentiation using a combination of large format 2D gel electrophoresis, image analysis, and mass spectrometry.
The basis for any analysis is the reproducibility of the results and the study design. Firstly, the reproducibility of large-format 2D gel electrophoresis was shown. Two samples of the same patient were analyzed using three replicate gels each. The spot quantitation of the two samples was found to be in good agreement. The relative mean standard deviation of the spot intensities within the replicate gels was 20% coefficient of variance. This allows us to analyze changes in the protein spot intensity that are smaller than a factor of two. The study design was optimized in order to account for technical and biological variation.
In the main study, 1800–2000 spots were quantified per gel. The large patient heterogeneity did not allow us to use a strict fold-change criterion for the selection of significantly changed spots between the two sample states. The variation of the spot intensity in one patient group was very much dependent on the nature of each individual protein. Therefore, a student’s t-test was employed to calculate the statistical significance for each spot. A total of 31 protein spots were found to be changed upon differentiation. Of these, 17 spots were unique for one of the samples, and another 14 spots were found to be highly significant (P = 99.9%). The effect of the Bonferroni correction and the false discovery rate is evaluated.