Back

Neonatal methylation-based predictors of childhood cognition

Smikle, R.; Mckinnon, K.; Vaher, K.; Turner, H.; Cruickshank, H.; Amir, R.; Chua, Y. W.; Conole, E. L.; Reynolds, R. M.; Batty, G. D.; Tsanas, A.; Murphy, L.; Whalley, H. C.; Marioni, R. C.; Cox, S. R.; Boardman, J. P.

2025-05-23 pediatrics
10.1101/2025.05.23.25328231 medRxiv
Show abstract

Preterm birth is closely associated with immune dysregulation in early life and subsequent learning and psychiatric disorders, but methods for stratifying infants at risk remain elusive. Epigenetic Scores (EpiScores) are relatively stable DNA methylation (DNAm)-based proxies of circulating proteins that can capture health-related exposures such as chronic inflammation. EpiScore of C-reactive protein (DNAm CRP) is associated with inflammatory burden in early life, atypical brain development following preterm birth (encephalopathy of prematurity), and adult cognitive ability. To evaluate the utility of EpiScores for predicting cognition in children born preterm, we investigated relations between 43 neonatal saliva-based EpiScores known to associate with low gestational age, and cognition assessed at 2 and 5 years of age in a cohort of 232 preterm and term-born children. DNAm CRP was negatively associated with 5-year Mullen Scales of Early Learning Composite (ELC) ({beta} = -0.273, p = 0.002). Association magnitudes were larger for children born earlier (DNAm CRP x gestational age, {beta}interaction = 0.181). EpiScores of CRTAM, NCAM1 and SLITRK5 were also associated with 5-year ELC in the full cohort (absolute {beta} range 0.219 to 0.267, Bonferroni-adjusted p-values <0.01). For preterm children, associations for DNAm CRP ({beta} = -0.318, p = 0.021) and DNAm CRTAM ({beta} = -0.307, p = 0.006) with 5-year ELC remained significant after adjustment for inflammatory exposures. We demonstrate associations between a range of neonatal salivary EpiScores and childhood cognition, suggesting the clinical value of EpiScores as early life markers of cognitive ability in children at risk of impairment warrants further investigation.

Matching journals

The top 6 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.