Key Findings

• Newborns who were later diagnosed with autism had lower levels of certain hormones, proteins and immune-related molecules in their cord blood.
• By using artificial intelligence (AI) and machine learning, the researchers built a model that could help predict which babies might later be diagnosed with autism.
• We observed certain protein disruptions persist into childhood (ages 7-10 years), suggesting we are looking at a core systemic aspect of autism biology that begins before birth.

Why is this important?

• Earlier support: Infants showing early biological signs could receive enhanced developmental surveillance.
• Improved understanding: Research into the biology of autism may help improve understanding of neurodevelopmental variation.

Congratulations to the research team; Aisling Noone, Kirsten Dowling, Daragh O’Boyle, Michael Carter, Anna V. Golubeva, Caitriona Scaife, Bodil H. Bech, Tine B. Henriksen, Louise Gallagher, Catherine Mooney, Ali S. Khashan, Deirdre M. Murray & Jane A. English.

Thank you to the funders:
This research was supported by the Irish Health Research Board (HRB) through an award to Dr Jane English.

Read more: https://doi.org/10.1038/s41380-025-03157-z

As part of the Springer Nature SharedIt initiative, the paper is also available at this link https://rdcu.be/eGN9d