Antibodies have been pin pointed as the main culprit in the reproducibility issue surrounding the research reagents sector today. However is it too cynical and perhaps plain simplistic to place all the blame on the antibodies alone? Could it be that it is not just the antibodies that are the issue, but the way scientists and manufactures handle them?
As the title suggests this blog will look at reasons other than antibodies that may have contributed to the reproducibility issue in this sector.
1) Wrong information in citations
There is no doubt that many scientists look to literature and in particular citations to find ‘good’ antibodies, and in spite of some publications describing successful use of a particular antibody -reproducibility is still pretty much unachievable. Many have attributed this to errors in publications or the lack of information provided by authors – something that can be easily rectified and could reduce at least some of the inconsistencies observed in experimental results.
2) OEM – a grey area
The google search engine is probably the most used platform to search for products on the web, but when searching for fit –for- purpose antibodies it may not prove to be the most effective search medium. This is partly due to companies using OEM (original equipment manufacturer) agreements, a common method used by companies to rapidly increase their product portfolio. Sometimes it can create a grey area within the market and the lack of information given about original manufactures makes comparing antibodies from different suppliers very difficult – especially if they’re depending on the first supplier names that appear in the search! For this reason, finding companies that produce their own antibodies and can provide up- to -date validation data is a big challenge for researchers who are not familiar with the biotech market. Websites such as CiteAb (http://www.citeab.com/) that help researchers find the right antibody and compare validation data are a lot of more representative of the biotech market.
3) “I wasn’t trained to validate antibodies , I was just trained that you ordered them” Rimm (Nature Baker May 2015)
The question of what makes a ‘good antibody’ still remains, as an antibody may perform excellently in one application but perform disastrously in another. Fridtjof Lund-Johansen from Oslo University (http://www.ous-research.no/home/blood/staff/6900) has developed a technology where many antibodies can be compared side to side – a high throughput kind of screening. He argues that without comparison it is difficult to differentiate the best antibodies – those that have a better signal or a stronger affinity than another antibody. Additionally, specificity of binding is also of great importance and could be tested by using knock-out samples within an assay for example. In other words without proper validation when using antibodies reproducibility will continue to be a problem.
It is apparent that the actual antibodies are not the real issue –when used correctly (without shortcuts!) they carry out their function very well. It is true, however, that without cooperation between biotech companies and researchers the reproducibility problem will be as prevalent as ever. Biotech companies can make certain that all of their antibodies are validated and make this data as readily and accurately available to researchers as possible. Scientists, on the other hand, need to be open to more innovative ways of validating antibodies and become savvier about the validation process and the requirements of their applications.