Biomarkers for Pediatric Sepsis Identified by Machine Learning and Bioinformatics

07 Apr 20264 min read0 viewsJournal Feed

GIST

Pediatric sepsis is a systemic inflammatory syndrome caused by dysregulated host immune responses, with a high mortality rate and a lack of effective biomarkers, posing significant challenges for early diagnosis and treatment. This study integrated six bulk RNA-seq datasets related to pediatric sepsis, including 497 patients and 116 healthy control samples.

Weighted gene co-expression network analysis was used to identify gene modules significantly associated with pediatric sepsis, and 237 high-confidence biomarkers were screened based on 14 machine learning models, among which RORA and GPR183 stood out in multiple models. Functional analysis indicated that these biomarkers were mainly involved in biological processes such as transcription and translation, the immune system, and cellular senescence.

Immune infiltration analysis revealed a significant reduction in adaptive immune cells such as B cells and CD8+ T cells and an increase in neutrophil and monocyte infiltration in pediatric sepsis patients, consistent with the “immunoparalysis” theory. Notably, RORA and GPR183 were positively correlated with CD8+ T cells, suggesting their potential role in regulating T cell function.

Additionally, we developed an open-source website for real-time application of biomarkers.

Clinical Editorial

Context and scope

The study addresses pediatric sepsis as a systemic inflammatory syndrome driven by dysregulated host immunity, highlighting the need for reliable biomarkers to aid early diagnosis and management.
An integrative approach merged six bulk RNA sequencing datasets from pediatric sepsis cohorts, comprising 497 patient samples and 116 healthy controls.

Analytical framework and biomarker identification

Weighted gene co-expression network analysis obtained gene modules with meaningful associations to pediatric sepsis.
A multi-model feature selection pipeline evaluated 14 machine learning models, yielding 237 high-confidence biomarkers.
Among these, two biomarkers—RORA and GPR183—emerged as prominent across multiple modeling approaches.

Biological interpretation and functional context

Functional enrichment implicated the biomarkers in transcriptional and translational processes, broader immune system activity, and cellular senescence pathways.
Immune infiltration profiling demonstrated a shift consistent with immunoparalysis: diminished adaptive immune cells (including B cells and CD8+ T cells) and increased neutrophil and monocyte presence in sepsis patients.
Notably, RORA and GPR183 showed positive associations with CD8+ T cell abundance, suggesting a putative role in modulating T cell function in this context.

Modeling validation and dissemination

An open-source web resource was developed to enable real-time application of the identified biomarkers.
A functional zebrafish model of sepsis was utilized, revealing downregulation of rora1, rora2, and gpr183 in diseased specimens, aligning with human biomarker signals.

Limitations and uncertainty

The summary reflects the reported associations and model-based findings; explicit diagnostic thresholds, performance metrics, and clinical outcomes are not provided in the available information.