I would like to know if anybody here knows of a method to detect the presence of ovalbumin--or any unique, egg-related molecule, in a baked good. Here I am anticipating that the "unique egg-related molecule" might alter under high temperatures, and so the suspect molecule for detection may or may not be any molecule which is present in a fresh, uncooked egg. I'm thinking of something that can be brushed on to a pastry to evolve a color change. Any help is appreciated.
There are a number of commercial kits available to determine the presence of egg in commercial products and they are based on a technique called sandwich ELISA (enzyme linked immunosorbent assay). A recent article in Food Research International by Gomaa and Boye discusses the impact of thermal processing on allergen detection (in this case casein, egg, gluten and soy). Egg recoveries with the kits are low, ranging from 48% down to 0. It appears that in some cases detecting egg in thermally processed samples is either very difficult to do or the egg products are transformed significantly that they can no longer be called egg (I did not read the paper in enough detail to determine if the authors speak to this point).
The kits tested in this study were from the following companies: Morinaga, Ridascreen and Ridascreen. I web search will reveal these products and additional information, but I'll leave that to the viewer as I do not want this question to be an endorsement or critique of any commercial products.
The alternative to these kits appears to be flow cytommetry, which has better recoveries, but does not fit the OPs conditions of economical for a mid-income business without access to a laboratory.
Egg white consists of mostly Ovalbumin. It is a protein, and so will be hard to distinguish from other proteins in a baked product. Egg white changes structure to a solid polymer on cooking. You could dissolve it in a mixture of detergent and hair perm, returning the Ovalbumin to its uncooked state, and then run it through a column to identify it.
There might also be biological assays you could run using enzymes/antibodies which would only interact with Ovalbumin.
I don't think there are any chemical means to identify a particular cooked protein easily.
Ask on Biology exchange?