That is an interesting question. I guess there is no definitive answer though, as the suitability of the different methods depends on various aspects, such as:
- What kind of changes in protein levels (number and magnitude) are you expecting? Since DiGE provides a visual impression of your samples and makes use of an internal standard, (small) changes might be picked up easier using this method. Of course, you will still have to excise spot(s) of interest to identify the corresponding protein(s).
- How much material do you have available for analysis?
A sensible DiGE experiment probably requires somewhat more material than an LC-MS-based experiment (although different labeling kits are available). And are you planning to use some kind of (isotope-)labeling for the LC-MS experiment too, in order to enable standardization? Are you considering to run multidimensional LC for the separation of your peptide mixture (assuming you'll be digesting your proteins first..)? And do you have the software to interpret either the DiGE or the LC-MS data properly?
- What biological fluids are you interested in? Do these require any depletion prior to analysis?
I think it comes down to the fact that the DiGE method will provide you a visual overview enabling better interpretation of potential differences (using the internal standardization), but requiring subsequent protein ID by mass spectrometry, while LC-MS of the peptides will tell you which proteins are in your mixture, but it will be more challenging to pick up any (subtle) differences between samples.