One of the main goals of shotgun proteomics studies is to identify by mass spectrometry as many proteins as possible, often from low-protein samples, in order to help answer important biological questions. All experimental steps need to be optimized to achieve a thorough proteomic analysis. One of the most important is an optimal extraction and solubilization of proteins through the use of detergents, chaotropes, and reducing agents. Although very efficient in solubilizing membrane proteins, the presence of these detergents, such as SDS, negatively affects the LC/MS analysis and limits protein identifications, and therefor they need to be removed before LC-MS analysis. Filter-aided sample preparation (FASP) is a generally accepted method for removal of detergents and chaotropes used for protein extraction as well as non-proteinaceous compounds such as salts, nucleic acids and lipids. Another critical aspect of proteomic analyses is the starting amount of protein in the sample. There is still little information about the efficacy of FASP method for protein-limited samples. The aim of this study was to compare two methods for SDS removal, ethanol precipitation versus FASP and we evaluate the effectiveness of FASP method using different filtration devices, as well as RIPA (0.1% SDS) and high-SDS (2%SDS) lysis buffers applied to low (1µg) and high (10-100µg) protein amounts.