Review 1: "Population Scale Proteomics Enables Adaptive Digital Twin Modelling in Sepsis"

RR:C19 Evidence Scale rating by reviewer:

• Potentially informative. The main claims made are not strongly justified by the methods and data, but may yield some insight. The results and conclusions of the study may resemble those from the hypothetical ideal study, but there is substantial room for doubt. Decision-makers should consider this evidence only with a thorough understanding of its weaknesses, alongside other evidence and theory. Decision-makers should not consider this actionable, unless the weaknesses are clearly understood and there is other theory and evidence to further support it.

***************************************

Review: This preprint provides evidence for the predictive value of a few proteins sampled at hospital admission to prognosticate and predict organ failure in patients with sepsis.

The authors perform feature selection to identify a set of proteins and design an 8d vector to assist with KNN-based classification of new patients, and demonstrate moderate predictive value. The notion that this approach is less prone to data distribution shift (or more "adaptive") needs additional validation. Additionally, the manuscript can benefit from additional comparative methods (e.g., a simple classifier) to demonstrate the superiority of the digital twin concept via KNN-based classification.

Overall, its an interesting works that demonstrates the potential value of enriching EHR datasets with additional biological data to better prognosticate patients with sepsis.