Objective To investigate the differential expression of proteins in the serum of diabetic patients quantitative proteomics technologies, and to identify specific protein biomarkers. Methods Serum samples from three diabetic patients and three healthy controls in Shenzhen in 2023 were collected. Immunoprecipitation and nano liquid chromatography-quadrupole-electrostatic orbitrap-linear ion trap mass spectrometry (easy-nanoLC-Q-orbitrap-fusion) were employed to perform fraction and no-fraction-based quantitative analysis of the serum proteome in diabetic patients and healthy controls. The mass spectrometry data were analyzed by MaxQuant1.6.1.0, and the differential proteins identified by the fraction-based method were subjected to gene ontology (GO) functional clustering and kyoto encyclopedia of genes and genomes (KEGG) pathway analysis. Results A total of 251 reliable proteins were identified by the no-fraction-based method, among which 17 differentially expressed proteins were selected. In contrast, 494 reliable proteins were identified by the fraction-based method, and 74 differentially expressed proteins were selected. Among them, seven proteins showed consistent changes in both methods. Five proteins were downregulated, including pregnancy zone protein (PZP), carbonic anhydrase 1 (CAH1), mannose-binding protein C (MBL2), fibrinogen alpha chain (FIBA), and haptoglobin (HPT). Two proteins were upregulated, including apolipoprotein(a) (LPA) and complement factor H-related protein 4 (CFH4). GO analysis revealed that the differentially expressed proteins were mainly involved in molecular functions such as glycosaminoglycan binding and heparin binding. KEGG pathway analysis showed that the differentially expressed proteins were mainly associated with complement and coagulation cascades, cell adhesion, PI3K-Akt signaling pathway, and other pathways. Conclusions Fraction-based no-fractional quantitative proteomics analysis provides an effective approach for analyzing differential protein expression in the serum of diabetic patients, and offers a valuable technique for screening specific protein biomarkers of serum samples in large-scale cohorts of disease.