Predicting Neonatal Sepsis

This episode explores a groundbreaking study published inb The Lancet that identified gene expression biomarkers capable of predicting neonatal sepsis before symptoms appear. Conducted in The Gambia, researchers analysed RNA sequencing data from healthy newborns to distinguish those who later developed early- or late-onset sepsis. Using machine learning, they identified a four-gene signature — HSPH1, BORA, NCAPG2, and PRIM1 — that accurately predicted early-onset sepsis at birth. The findings also revealed that early sepsis profoundly alters immune and metabolic development in the first week…