Marissa D Pokharel Anlin Feng Ying Liang Wenli Ma Saurabh Aggarwal Hoshang J Unwalla Steve Black Ting Wang ^*

• Florida International University, Miami, United States

Sepsis is a severe and life-threatening medical syndrome that can lead to organ failure and death. Despite advances in medical treatment, current therapies are often inadequate, with high septic mortality rates. Therefore, there is a critical need for reliable prognostic markers to be used in clinical settings to improve the management and outcomes of patients with sepsis. Recent studies have suggested that mitochondrial dynamics, including the processes of mitochondrial fission and fusion, are closely related to the severity of sepsis and the status of inflammation. By monitoring transcriptomic signals related to mitochondrial dynamics, new and reliable biomarkers can be engineered to more accurately predict sepsis survival risk. Such biomarkers would be invaluable in clinical settings, aiding healthcare providers in the early identification of high-risk patients and improving treatment strategies. To achieve this goal, we utilized the major mitochondrial fission regulatory protein dynamin-related protein 1 (Drp1, gene code DNM1L) and identified Drp1associated genes that are enriched with sepsis survival genes. A 12-gene signature (GS) was established as a differentially expressed gene (DEG)-based GS. Next, we compared genes of proteins that interact with Drp1 to sepsis survival genes and identified 7 common genes, establishing a GS we term as protein-protein interaction (PPI)-based GS. To evaluate if these GSs can predict sepsis survival, we used publicly available human blood transcriptomic datasets from sepsis patients. We confirmed that both GSs can successfully predict sepsis survival in both discovery and validation cohorts with high sensitivity and specificity, with the PPI-based GS showing enhanced prognostic performance. Together, this study successfully engineers a new and validated blood-borne biomarker (PPI-based 7-gene GS) for sepsis survival risk prediction. This biomarker holds the potential for improving the early identification of high-risk sepsis patients and optimizing personalized treatment strategies to reduce sepsis mortality.