For proteome analysis, the most important criteria to be considered when evaluating methods of protein detection are limit of detection, dynamic range, homogeneity, and compatibility with MS analysis. The properties of some of the commonly techniques are shown. Most proteomics studies require limits of detection in the low nanogram protein range, which would correspond to the femtomole levels usually detectable by MS analysis (eg, 1-20 ng, corresponding to 100 fmol of a 10-200 kDa protein). A wide linear dynamic range is essential when the aim of the study is relative quantification between samples. The homogeneity of the detection refers to the uniformity of staining between different classes of proteins. When the method targets only one amino acid, proteins that do not contain that particular residue are not detected (eg, cysteine). An ideal detection method would target all amino acid residues equally, so that the signal detected would be directly proportional to the length of the amino acid chain. The compatibility of the method of detection with MS analysis is particularly important for the recognition of the proteins. The presence of formaldehyde or glutaraldehyde in various solutions used for silver staining (fixing, silver impregnation, and developing) interferes with protein recognition because these aldehydes alkylate the α and ϵ amino groups of proteins.¹⁰  Silver staining methods compatible with MS analysis were developed by removing the glutaraldehyde and decreasing the use of formaldehyde in the stain.⁹⁹  Although most detection methods allow only intergel comparison, the new generation of cyanide dyes that chemically modify lysine or cysteine residues offer the potential for multiplexing in gel-based experiments (see “Separation procedures and gel electrophoresis-based quantification for proteomics” and Figure 2) while not compromising downstream MS analysis. Regardless of the detection method, the principal method for relative quantification is densitometry analysis. Providing that there is a good reproducibility between the 2-DE protein patterns, densitometry is an important tool in proteome analysis because it allows the simultaneous detection and quantification of hundreds of proteins from a multitude of cell states. A number of computer-assisted image analysis software packages that allow the comparison of 2-DE protein patterns, such as PDQuest (Bio-Rad, Hercules, CA), Melanie (Bio-Rad), and Progenesis (Non-linear Dynamics, Durham, NC), are now commercially available. However, the technique suffers from a number of limitations, either inherited from 2-DE's lack of reproducibility and poor dynamic range of staining methods, or, depending on which software is used for the analysis, from its own poor sensitivity of detection. New software and methods constantly improve protein detection and image warping, but the success of the matching still depends extensively on the similarity of the protein patterns. For in-gel comparison, DIGE experiments plus the appropriate analytical software therefore offer an effective way forward in relative quantification of the hundreds of proteins found in 2-DE gel experiments.