Prediction of early Toxoplasma gondii infection during pregnancy combining host biomarkers and machine learning

A four-biomarker serum panel (CXCL9, CXCL10, IFN-{gamma}, and serum amyloid A) predicts early Toxoplasma gondii infection in pregnant sheep with over 74% accuracy and 100% sensitivity in natural abortion cases.

• Why it matters: Early congenital toxoplasmosis in sheep can cause early fetal death during days 7–14 post-infection, when standard diagnostics are ineffective, and better biomarkers are needed for timely detection.
• What they did: Serum from 22 experimentally challenged pregnant sheep and 6 uninfected controls was profiled for 27 biomarkers, then machine-learning models were tested against sera from infections with Neospora caninum and Chlamydia abortus and validated using 19 natural abortion outbreaks.
• The result: The panel specifically detected early infection with over 74% accuracy and achieved 100% sensitivity in natural infection samples, supporting rapid diagnostic development for livestock and potential translation to human diagnostic gaps.

AI summary of the abstract · may contain errors · not medical advice · read the paper.