I'd like to estimate the approximate number of species that can be separated by a LC-MS system without too much overlap. If raw mass spectra are processed to peak lists, so that each species is represented by one mass (each mass with the same proportional-to-mass error), rather than m/z values for a number of isotopic and/or charge state peaks, it should be easy to estimate a peak capacity. For example, when analyzing intact proteins using FTICR-MS, numerous isotopic peaks from numerous charge states can be processed to a single mass with a ppm error; as isotopic resolution (and the best mass measurement accuracy) is limited to a certain mass, the maximum number of peaks that could fit between a minimum mass and this maximum can readily be calculated. This number would be analogous to a chromatographic peak capacity and could be multiplied by the on-line separation peak capacity to give a LC-MS system peak capacity. The logic outlined in Davis and Giddings, Anal. Chem. 1983, 55, 418-424, could then be applied to determine how many species could be analyzed with a reasonable confidence of getting single-component peaks. My question is, is this a reasonable idea? Am I missing something important?