4. What do you know about your protein or epitope of interest? Where is it located? Is it subject to post-translational modification (PTM)? Identify whether you want an antibody that will recognize the whole protein, a certain segment of it (e.g., N- or C-termini), or a specific peptide sequence within the protein (e.g., a phosphorylation motif). Check immunogen details to be doubly sure your antibody will tell you what you want to know.
5. Bear in mind the species you’re working with; if an antibody was raised in the same species as your target one, it may cause problems later at the detection stage. There are ways around this if unavoidable, though, such as including normal serum from the species in which the secondary antibody is raised in your blocking buffer.
6. Look at the validation data on each vendor’s website, and check their depth and quality. Is there simply verification of antigen recognition by ELISA testing, or are there Western blots and immunohistochemistry (IHC) data, too (as well as any other validations)? What sorts of samples were used in these tests? Make sure you’re viewing data obtained from whole cell lysates and whole tissue preparations as opposed to those prepared with purified recombinant proteins, which is usually not an accurate reflection of how the antibody will perform in the lab.
7. The above being said, if your application is missing validation data this does not necessarily mean the antibody will be unsuccessful; you still have a few options. Does the company offer trial samples (unlikely but not unheard of) or, better, do they cover their products with a money-back guarantee? (You can try larger volumes, confident that you’re not wasting any of your grant money with the latter option.) Money-back guarantees should be of the ‘no-quibble’ kind (make sure the company you buy from doesn’t require reams of data for you to qualify) and the best ones offer you the option of your money back as opposed to credit.
8. If unsure of your application, polyclonal antibodies raised against the full-length protein—sometimes referred to as whole-protein antibodies—will cover most of your needs in Western blot, IHC, and immunoprecipitation (IP) experiments. They are ideal in the case of the latter two as this type of antibody recognizes more of its target and three-dimensional conformations—especially handy for IP!
9. There is something to be said for seeking out smaller vendors and original producers rather than the larger antibody ‘supermarket’. Companies who produce their own antibodies know them inside and out and can detail everything about an antibody’s production process. They can also often perform validations at your request and can send you hard copies of their original validation data. They are also more likely to offer comprehensive custom production services in the event you are unsuccessful in your search for a pre-made antibody against your chosen target.
10. With experimental antibodies* (those about to undergo validation), contact the company to see if a trade is possible—your future validation results in exchange for a free vial. If you can give something back to the company, even if it’s negative feedback, it is seen as some compensation for the trial, and if you do get positive data, the situation is win-win.