Back

Biological landscape of acute illness in children in sub-Saharan Africa and South Asia

Mudibo, E. O.; Sande, C.; Ngari, M. M.; Jenkins, B.; Singa, B. O.; Lancioni, C. L.; Diallo, A. H.; Mbale, E.; Mupere, E.; John, M.; Bamouni, R. M.; Prendergast, A. J.; McGrath, C. J.; Koulman, A.; Bandsma, R. H. J.; Tickell, K. D.; Walson, J. L.; Berkley, J. A.; Gonzales, G. B.; Njunge, J. M.

2025-09-03 systems biology
10.1101/2025.09.01.672585 bioRxiv
Show abstract

Childhood illnesses including pneumonia, diarrhoea and malaria are leading causes of hospitalisation and mortality in resource-limited settings. However, we lack understanding of whether systemic responses to such diverse clinical syndromes are shared or specific, how they are impacted by malnutrition and how they differ from well children. We performed multi-omic profiling of plasma proteins, and serum metabolites and lipids in acutely ill hospitalised and well children in sub-Saharan Africa and South Asia. Using network-based clustering and mixed-effects modelling, we identified common and syndrome-specific omics responses to acute illness. We found that malnutrition often modifies host responses to disease. Although the internal structure of individual omics modules was largely preserved between ill and well children, the interactions between these preserved modules were markedly reorganised during acute illness. Compared to well children, biological systems in hospitalised children were more interconnected, exhibiting denser cross-omics interactions. These findings reveal widespread multisystem mobilisation during paediatric acute illness, offer deeper mechanistic insights and highlight candidate pathways for therapeutic intervention in high-burden settings.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.