Machine learning model to identify gut microbiome-derived metabolites as potential biomarkers of autism spectrum disorder: a pilot study

This pilot study uses machine learning to analyze theoretical metabolite production from gut bacteria in ASD versus healthy controls, achieving 85% classification accuracy with results consistent with prior research. The identified metabolites (amino acids, volatile organic compounds, short-chain fatty acids) align with known mechanisms affecting neurotransmitter synthesis and gut barrier function in ASD.