Links between gut microbiota and autism spectrum disorder (ASD) have been explored in many studies using 16S rRNA gene amplicon and shotgun sequencing. Based on these links, microbiome therapies have been proposed to improve gastrointestinal (GI) and ASD symptoms in ASD individuals. Previously, our open-label microbiota transfer therapy (MTT) study provided insight into the changes in the gut microbial community of children with ASD after MTT and showed significant and long-term improvement in ASD and GI symptoms. Using samples from the same study, the objective of this work was to perform a deeper taxonomic and functional analysis applying shotgun metagenomic sequencing. Taxonomic analyses revealed that ASD Baseline had many bacteria at lower relative abundances, and their abundance increased after MTT. The relative abundance of fiber consuming and beneficial microbes including Prevotella (P. dentalis, P. enoeca, P. oris, P. meloninogenica), Bifidobacterium bifidum, and a sulfur reducer Desulfovibrio piger increased after MTT-10wks in children with ASD compared to Baseline (consistent at genus level with the previous 16S rRNA gene study). Metabolic pathway analysis at Baseline compared to typically developing (TD) children found an altered abundance of many functional genes but, after MTT, they became similar to TD or donors. Important functional genes that changed included: genes encoding enzymes involved in folate biosynthesis, sulfur metabolism and oxidative stress. These results show that MTT treatment not only changed the relative abundance of important genes involved in metabolic pathways, but also seemed to bring them to a similar level to the TD controls. However, at a two-year follow-up, the microbiota and microbial genes shifted into a new state, distinct from their levels at Baseline and distinct from the TD group. Our current findings suggest that microbes from MTT lead to initial improvement in the metabolic profile of children with ASD, and major additional changes at two years post-treatment. In the future, larger cohort studies, mechanistic in vitro experiments and metatranscriptomics studies are recommended to better understand the role of these specific microbes, functional gene expression, and metabolites relevant to ASD.