ABSTRACT
Proteomic analyses have benefi ted enormously from advances in chromatography and mass spectrometry (MS), which together can provide a vast amount of data from the myriad of components present in typical samples. The daunting challenge in proteomics is to generate useful information from these data regarding the presence and abundance, either in absolute or relative terms, of the potentially thousands of proteins found in a biological system. With the recent shift from genomic to proteomic studies has come the realization that proteomics is far more complicated than the approximately 30,000 genes in the human genome would predict due to the multiplicity of protein structures stemming from a single gene.1,2 Mammalian proteins are subject to a variety of posttranslational modifi cations, with over 200 different modifi cations being reported.3 These modifi cations may be combinatorial as is often the case with glycosylation4,5 and phosphorylation,6-10 with a multiplicity of structure resulting from heterogeneity or incomplete occupation of one or more of the sites. Amino-terminal protein processing and acetylation produces additional structural variation from genome predicted protein sequences.11-14 Each of these gene product modifi cations may have important physiological implications that result in altered biological activity, often as a normal consequence of cellular physiology, including mechanisms for regulating activity, interactions, cellular transport,15-17 and protein lifetime.18 Of interest to many proteomic researchers, such changes may also occur as a result of cellular insult or disease processes.19-23
Global proteomics studies are further complicated by the large dynamic range of protein concentrations present in a biological system. In a typical cell this range is 106,while in biofl uids such as serum,24,28 the dynamic range can be as high as 10. 11 This far exceeds the dynamic range accessible for any current MS detection technique, and hence limits the detectability of low concentration proteins in the presence of more highly abundant ones. In fact, detection of highly abundant proteins is often so ubiquitous that response from low abundance components is completely masked.
