Abstract:OBJECTIVE To uncover critical active proteins influencing sepsis outcomes through multi-omics analysis. METHODS This study collected peripheral blood from sepsis patients (NS = 26, SV = 27) and controls (Con = 16). Cellular heterogeneity was assessed using scRNA-seq. Cellular populations were identified through clustering and annotation. GSVA was employed to detect pathway alterations in sepsis, while the Viper algorithm estimated protein activity at the single-cell level. Signaling networks were investigated via cell-cell communication analysis. Differentially expressed proteins were identified by DIA proteomics and confirmed through integrated analysis. Prognostic value was evaluated via meta and survival analyses. RESULTS scRNA-seq of 22,673 features within 34,228 cells identified five cellular clusters and 253 active proteins via Viper, validated by DIA (FC > 2, P < 0.05). Four proteins (SPI1, MEF2A, CBX3, UBTF) with prognostic significance were discovered and mapped onto the cellular landscape. GSVA enrichment analysis revealed that the NS group exhibited significant alterations in pathways related to cellular apoptosis and inflammatory responses, while the SV group displayed increased activity in DNA repair and cellular survival pathways. CONCLUSION The study's findings advance the understanding of sepsis pathophysiology by linking differentially active proteins to patient prognosis, paving the way for targeted therapeutic strategies.

Novel Active Proteins for Sepsis Prognosis Revealed through ScRNA-seq and Quantitative Proteomics.