Reproducibility Study - Phase I
Full details and results are being submitted for publication and will be available soon. But you can see initial data from this groundbreaking study showing that protein expression analysis based on 2D PAGE is reproducible across-labs and highly reproducible within-lab.
 |
Slides presented at the Workshop on Proteomics Validation, HUPO 2008 7th Annual World Congress. |
 |
Poster on Reproducibility of 2D gel-based proteomics experiments shown at 8th Siena Meeting, August 2008. |
Some of the data has been represented below to give you more detailed explanations of the important figures. You can download the protocols used in the study to try in your own laboratory.
 |
 |
| Whole image pixel-based Pearson correlation coefficients for intra-lab datasets (across all gels including the treatment differences) show very high reproducibility. | Intra-lab technical replicates show a very high level of reproducibility and linearity at the spot level. |
 |
|
| Inter-lab technical replicates show a high level or reproducibility and linearity at the spot level. |
Similarity between the top 50 differential spots
Using fully automatic analysis of 2D PAGE images increased reproducibility compared to user-editing, even by experts. The values show the similarity (%) of the top 50 differential spots from each image set following analysis with user-editing and with fully automatic analysis compared to a constrained analysis from a professional analyst.
| Unconstrained User Analysis |
Fully Automatic Analysis |
|
|---|---|---|
| User1 | 92% | 98% |
| User2a | 84% | 98% |
| User2b | 76% | 100% |
| User2c | 88% | 100% |
| User3 | 70% | 88% |
| User4 | 88% | 84% |
| User5a | 90% | 94% |
| User5b | 94% | 82% |
| User5c | 94% | 94% |
Want to find out more?
Proteomonitor article - Trying to Settle Ongoing Dispute, HUPO's IAB Finds 2D Gel Experiments are Reproducible
The HUPO Reproducibility Study introduced by Jan Van Oostrum, one of the Fixing Proteomics founders
