Antibodies against common cellular targets (e.g. ubiquitin) are available from many different commercial sources. But are they all created equally? And are they appropriately validated to demonstrate specificity? This is important, because poorly characterised antibodies can yield unreliable results, and non-reproducibility between different labs.
On the one hand, there is cause for concern that antibodies do not live up to the standards that are claimed – simply, they can’t be relied upon to detect the right protein.
But on the other, it is essential to have a robust method for ensuring that antibodies are validated appropriately – and that means using protein samples with appropriate controls (e.g. a knock out mutant and an overexpressor of the protein in question). Without this, antibodies can’t really be given a fair trial.
I reckon caveat emptor applies – as a thorough scientist, it is in your interest to ensure that you do not trust blindly that an antibody does what it says it will do. You have to convince yourself, using experimentation, if you hope to convince others. Meanwhile, initiatives such as Antibodypedia are promising ways to share validation data for proven antibodies. Will plant science catch up?