Background: The COVID-19 pandemic presented a challenge to biological researchers and clinicians involved in patient diagnosis and treatment due to an increase in SARS-COV-2 mutations. This demonstrated the necessity of a thorough molecular understanding of the disease to categorize various patient groups. Methods: LC–MS/MS-based untargeted proteomics analysis was performed to profile the changes in saliva proteins of COVID-19 patients with different laboratory results. Herein, we performed differential proteomic analysis of RT-PCR negative saliva samples with two groups of patients- (1) TG(test group)1: RT-PCR positive and antibody positive (2) TG2: RT-PCR positive and antibody negative. Statistical analysis: Statistical analysis was performed using a web-based program, MetaboAnalyst 5.0. Results: In total, 2784 proteins were identified in analysed saliva samples. Compared to RT-PCR negative samples, a total of 62 and 372 proteins were found differentially expressed in TG1 and TG2 groups, respectively. Aminopeptidases and clathrins, which are both involved in viral entry, were found upregulated in both test groups. Down regulated proteins were abundant and included proteins like aminotransferases, which could be attributed to activation of alternate metabolic pathways. PLS- DA analysis based on protein expression clearly separated both test groups from controls. However, separation was not as robust when the test groups were compared with each other. Conclusion: Differential expression of proteins has the capability of classifying different COVID-19 patient groups.