51 types of bacteria, 18 viruses, 14 archaea, 7 fungi, and a dozen metabolic pathways, were altered in autistic children. - with machine learning autistic children identified up-to 82% accuracy, based on 31 microbes and bio functions in digestive system