Computational analysis of high-density peptide microarray data with application from systemic sclerosis to multiple sclerosis.

Auto-antibodies are implicated in the pathophysiology of various autoimmune diseases. High-density peptide microarrays incubated with human serum can detect antibody reactivities against thousands of peptides. This enables the identification of new auto-antigens and the determination of the parts of protein antigens (epitopes) that are recognized by antibody paratopes. We discuss the utility of peptide microarrays to investigate epitope-antibody-recognitions (EAR) from systemic sclerosis to multiple sclerosis. The technology can help to establish reliable diagnostic and prognostic biomarkers employing a combination of antigenic peptides. We describe the specifics of peptide microarray data and present bioinformatic methods for their analysis. Quality control, data pre-processing and the filtering of specific peptides are demonstrated on an example data set. Peptide microarrays representing 24 selected proteins by 3235 overlapping 15mer peptides were used to measure antibodies in serum of 10 patients with limited cutaneous systemic sclerosis (SSC) and 10 healthy blood donors. The data showed a sparse and skewed distribution, and we observed strong individual differences since many peptide sequences were bound by antibodies of only one serum sample. In the sera of the SSc patients, but not of the healthy controls, we found antibodies to three peptides MGPRRRSRKPEAPRR, TPTPGPSRRGPSLGA and GPSRRGPSLGASSHQ that share a similar sequence motif (GP-R/S-RR). These peptides map to two known linear epitopes at the N-terminus of centromere protein A (CENPA), demonstrating the utility of peptide microarrays. Presented experimental and bioinformatic approach can be applied in the same manner for multiple sclerosis research.