Gain confidence in running high quality 2D gels

The goal of HUPO Reproducibility Study Phase II project was to facilitate the introduction and use of 2-DE by providing reference protocols, images and samples. This work describes the experiences with those reference materials, which were distributed to around 20 labs worldwide with the request to produce at least three gels and submit the images. These were compared to a set of gold "standard" reference images run with high stringency using the same protocol and sample¹.

Figure above below shows Gold Standard gel image (A) compared to a gel image from a participatory lab (B). Study participant followed the protocol and generated the image that is compared to the Gold Standard. Highlighted region in the Red Box is for visual reference.

Summary results and conclusions

Thanks to all the people who took part in this study. Using the protocol and with support from the study co-ordinators² we showed,

Intra-lab Reproducibility – Highly reproducible for gels run within the same lab
Inter-lab Reproducibility – 10/17 labs (~60%) were able to generate gel images which fall within the 95% confidence bound as determined by PCA

You can read initial results presented at the HUPO 2009 meeting in Toronto and follow up article that highlight the aims and findings of the study. A manuscript is in progress, due for submission in early 2010.

Initially several participants had difficulties reaching an optimal result with the reference sample. In several cases, the observed inconsistencies immediately led to suggestions towards improvement.

Two images from the same study participant (A) Attempt 1 (B) Attempt 2 after feedback from expert analyst. This illustrates the value of consultative approach leading to the improvement in the quality of the image. Highlighted region in the Red Box is for visual reference.

This result illustrates that a complex proteomics method such as 2-DE can benefit greatly from the availability of reference materials. With the immediate feedback on the quality of their gels, users can gain confidence in being able to generate reproducible data with their precious samples. This work also showed,

Use of same starting material will be helpful
Availability of reference images (intra and inter-lab) could reduce bias
Even the best labs will see user-dependent variation
Minimize variation arising from product quality
Iterative approach will help reduce issues - streaking, poor spot resolution and artefacts

Gel QC analysis

Tools and approaches from the study are available to read about and download from the Fixing Proteomics Campaign website. These help you to 'calibrate' your own 2D gel running setup and make your proteomics results reproducible across-labs.

The software we used to perform the 2DE gel image QC analysis is being developed into a "user-friendly" workflow that anyone can use. It'll be free-to-use for the proteomics community but you can register for a preview. For more details about the Gel QC analysis you can download our two supplementary information sheets, 2D gel QC workflow and PCA explanation.

¹ Gold Standard Reference 2D gels and 2D gel images provided by Dr. Hans Voshol, Novartis Institutes for BioMedical Research, Switzerland

² Study co-ordinators were Bio-Rad, Nonlinear Dynamics, CILBiotech and