Antibody validation is critical to reproducibility and a necessary step to increase the security of antibody specificity in every specific context relating the target protein, sample, and application. Yet all too often, antibody-based published data cannot be reproduced. If you want your scientific project to be taken seriously, you should spend some time upfront to validate your antibodies before starting the project. Read on to know why this is so important.
Antibodies are among the most common research tools routinely used for a variety of assays such as immunohistochemistry and immunofluorescence, western blot, immunoprecipitation, enzyme-linked immunosorbent assays and more.
Much has been written about the role of antibodies in data irreproducibility. Nevertheless, a recent analysis estimates the presence of flaws or errors in every step of preclinical experiments.
Four categories of irreproducibility have been found: from the study design to laboratory protocols, from data analysis to biological resources that include animal models, cells lines and, of course, antibodies1.
Figure 1. The magnitude of the reproducibility crisis and key sources of irreproducibility (1, Adapted).
Lack of highly validated antibodies comes with great consequences
Although antibodies are not the leading causes of lack of data reproducibility in preclinical research, the spread of poorly validated, hence poorly performing, antibodies unquestionably harm the credibility of research data.
According to the United States Food and Drug Administration (USFDA) “validation is the process of demonstrating, through the use of specific laboratory investigations, that the performance characteristics of an analytical method are suitable for its intended analytical use”.
Antibody validation is critical to reproducibility, and it is both costly and time-consuming. However, it is a necessary step to increase the security of antibody specificity in every defined context relating the target protein, sample, and application. Yet all too often, antibody-based published data cannot be reproduced.
Currently, there is no consensus and no global standard for antibody validation and for the reference data that must be provided in journal publications. Consequently, crucial specific protocol information is often unclear, misleading or entirely unavailable. The evidence shows that 54% of biomedical publications have inadequate descriptions of study design and analytical methods; and, about 85% of worldwide research funds are wasted every year due to this issue2-4.
Research is not a game of chance
Ambiguous bodies of knowledge have grown from using unspecific antibodies, that escalate into misdirected studies and costly efforts by labs and companies; not to mention clinical trials run inadequately.
One example of such negligence is described in a case study published a year ago in the journal Nature Communications. A Sweden-based research team led by Professor Cecilia Williams reported that the reliance on insufficiently-validated antibodies may have led cancer research down a 20-year dead end. For two decades researchers have sought a way to target an estrogen receptor in the hope they could improve breast cancer survival.
Unfortunately, the target receptor did not actually appear to be at the expected location. The research team said their study invalidates all but one of the antibodies used to detect estrogen receptor beta (ESR2) in breast cancer tissues. These antibodies have instead been mistaking other proteins for ESR2, so, the data generated and published using these antibodies is not trustworthy5.
Resetting the (mal)practise
This malpractice has received much attention in the past few years and is now slowly reversing. In fact, due to the recent intensified publications on the crisis of reproducibility, there is a growing attention to the quality of commercial research antibodies, in the whole scientific community.
In order to boost the reliability in laboratory data, scientific journals have strengthened their submission criteria and toughened their editorial policies, making it a requirement for researchers to report experimental protocols more thoroughly and encourage the use of proven validated antibodies. Having this information available not only saves you time and effort, it also makes it easier for other authors to replicate your data with greater consistency across different labs and purposes6.
What does that mean for you?
This renewed attention in reporting key methodological details and transparency is undoubtedly an improvement. But it is still a small step. Journals editors and antibody suppliers must be accountable. However, in the end, it is your responsibility, as a researcher at the bench, to spend some time evaluating the best antibody that suits your needs. From your perspective, it means you must carefully choose the antibody, proven to be specific, selective, and reproducible in your context, such as sample and application, for which you will use it.
Now tell me: how robust is the evidence behind your antibody?
Take a look at our post "5 tips about validation when choosing an antibody" where we talk about the approved international guidelines for antibody validation.
1. Freedman LP, Venugopalan G and Wisman R. Reproducibility2020: Progress and priorities F1000 Research 2017, 6:604
2. Ioannidis JPA. Why most published research findings are false. PLoS Med. 2005; 2:696–701.
3. Macleod MR, Michie S, Roberts I, et al. Biomedical research increasing value, reducing waste. Lancet. 2014; 383:101–104.
4. Vasilevsky NA, Brush MH, Paddock H, Ponting L, Tripathy SJ, LaRocca GM, Haendel MA. On the reproducibility of science: unique identification of research resources in the biomedical literature. PeerJ 2013; 1:e148
5. Andersson S, Sundberg M, Pristovsek N, Ibrahim A, Jonsson P, Katona B, Clausson CM, Zieba A, Ramström M, Söderberg O, Williams C, Asplund A. Insufficient antibody validation challenges oestrogen receptor beta research. Nature Commun. 2017; Nov 29;8:16164.
6. Giraldo O, Garcia A, Corcho O. (2018) A guideline for reporting experimental protocols in life sciences. PeerJ 6:e4795