I am concerned about the specificity of this antibody if it "works great in IF," but gives you results that you don't like in western. What leads you to believe that it works great in IF? Because it gives a staining pattern that looks interesting? Or are there other data that lead you to believe that it is faithfully reporting the expression of your protein?

Pre-incubating with antigen is an okay control... but the gold standard, if you want to be truly confident in your Ab is doing the staining in a knockout and seeing your pattern disappear. You mention that there is no mutant, but certainly you can use a morpholino to deplete your protein of interest and repeat the staining. If you and your PI decide to use pre-incubation as the only specificity control, I'd aim for 1:1 stoichiometry. Determine your amount of antibody and antigen in ug, then use kDa to convert to moles and combine equal amounts.

An antibody that does not work well in Western blots may be one that recognizes a structural epitope rather than a contiguous sequence in the protein. You still will have to prove that it recognizes the correct antigen with good affinity and specificity, but will need to use methods that leave the protein intact. Depending on the properties of your antigen, you could, for example, test whether your antibody is able to pull down a protein of the expected size from a cell extract, e.g. after gentle solubilization with detergent if you are looking for a membrane protein. If you have the purified antigen, you can also test your antibody in an ELISA setting.

Ed Van Veen Thank you for your suggestions. We will try MO injection.

Annemarie Honegger Thank you so much. we will try!