Study on saliva proteome and bioinformatics in patients with chronic gastritis

Objective To explore protein expression profile in chronic gastritis saliva by proteomics mass spectrum techniques, and seek for the specific biomarkers of diagnosis for chronic gastritis. Method Peptide mass fingerprint of saliva were collected from chronic gastritis patients and healthy subjects by using MALDI-TOF-MS technique after saliva sample had been treated with WCX magnetic beads. The diagnostic cast was developed based on the peptide mass fingerprint which were obtained from the saliva of chronic gastritis patients and healthy subjects. Result Totally 74 protein peaks were identified in the chronic gastritis group and the control group. 5 specific protein peaks (P <; 0.05) were found with statistically significant differential expression level. The 3 specific protein peaks with a mass-to-charge ratio (m/z) of 5502.36Da, 1441.75Da and 3442.47Da were used to build a predictive model for the diagnosis of chronic gastritis. The predictive model has identification rate of 91.7% and predictive ability of 73.33%. Clinical back substitution analysis indicated that this diagnostic model can discriminate chronic gastritis from controls with a precision rate of 90.63% (29/32), a sensitivity of 100% (14/14), a specificity of 83.33% and a positive predictive value of 82.35% (14/17). Conclusion Saliva protein fingerprint mass spectrum was initially obtained, which established a certain base for future ulterior proteome studies in chronic gastritis. Furthermore, the diagnostic cast on 5502.36Da, 1441.75Da and 3442.47Da protein peaks of protein expression mass spectrum from chronic gastritis saliva protein was developed to discriminate chronic gastritis clinically. A completely new method and idea of chronic gastritis proteome investigation was provided, which has a really important significance on screening the specific biomarker and needs to be further studied.