Sepsis is an underestimated killer. Nearly a quarter of patients treated for sepsis in hospital will die, but because so many different illnesses can predispose patients to experiencing it, it's overlooked as a direct cause of death. Yet approximately 20% of deaths worldwide are caused by sepsis, and currently we have no treatments that tackle it directly.

One of the reasons it’s so hard to understand and treat sepsis is that it is multifaceted. Sepsis arises when the immune system fails to control an infection and malfunctions, causing multi-organ failure. Many different infections can cause sepsis, and its symptoms and progression vary between patients and over time in the same patient. Its early symptoms are similar to those of many other illnesses, which makes it difficult to diagnose quickly and initiate timely treatment, contributing to high mortality.

Systems immunology offers a potential solution to this diagnosis problem by using mathematical and computational modeling to study the immune system in the context of all the body’s other systems. It does this by using different types of clustering analysis to identify patterns in large volumes of omics data, ranging from transcriptomic data (what genes show altered expression) to proteomic and metabolomic data—data that tell us about the body’s reaction to its physical circumstances, in this case sepsis, in incredibly fine-grained detail.