Antibody profiling is becoming increasingly important for the development of new diagnostic tests that enable early detection of disease. The high abundance of antibodies and their relative stability make them excellent biomarker candidates, especially for autoimmune diseases. In addition, the identification of disease related antibodies may provide insight into disease pathogenesis and facilitate the development of antigen-specific therapies. We have employed bacterial cell surface display to develop a novel method that can be applied to identify disease related antibodies in serum. As a test case, an antibody with known specificity (anti-T7 tag) was spiked into pooled human IgG. A bacterial display library expressing random 15 amino acid peptides was used in conjunction with two-color fluorescence activated cell sorting (FACS) to identify peptide mimitopes that cross-react with the IgG. After three rounds of FACS, the library was successfully enriched for bacteria displaying peptides that bind to the target antibody as confirmed by DNA sequencing. The use of two-color FACS allowed for quantification of enrichment at each round of sorting as well as rapid isolation of bacteria expressing the antibody epitope. This method can enable the discovery of prognostic serological markers from the circulating antibodies in serum as well as the epitopes that may be used to detect them. The application of this method to antibody profiling in a mouse model of atherosclerosis will also be discussed.