Table 1.

Characteristics of included studies.

Characteristics							Quality assessment by the NOS tool§			Summary of findings
										N. of subjects		Results
First author, year	Cases	Controls	Type of sample	Nucleic acids extraction	Computational pipelines	Potential confounders	Selection	Comparability	Outcome/Exposure	n. cases	n. controls	Data
GUT MICROBIOMA
Prehn-Kristensen (32)	Males with ADHD diagnosis according to the DSM-IV -TR criteria	Male healthy children recruited through newspaper announcement	Stool samples: collected in fecal collection tubes and stored at 4°C until preparation	Total DNA was extracted using FastDNATM KIT FOR SOIL	MOTHUR	Ten children with ADHD had been taking medicine for more than 1 year to treat ADHD symptoms. Nine of them discontinued medication for at least 48 h prior to sample collection	*		*	14 (M = 14)	17 (M = 17)	Alpha Diversity Index Chao1: no differences Shannon: ADHD < HC (p = 0.036) Observed: no difference. Correlations with alpha diversity Hyperactivity: R = −0.35 (p = 0.03) Impulsivity: non-significant Attention: non-significant
Jiang (33)	ADHD diagnosis according to the DSM-IV criteria	Neuro-typical (NT) control enrolled via advertisements	Stool samples: collected by parents in sterile plastic cups and stored at −20°C at home. Samples were kept in an icebox that was delivered to the laboratory within 30 min, and then stored at −80°C.	Fecal microbial DNA was extracted from 200 mg of feces using the QIAamp DNA Stool Mini Kit	QIIME (version 1.7)	Children who were taking probiotics or antibiotics during the 2 months prior to the fecal sample collection, who had apparent gastrointestinal symptoms, or had been or were currently taking medications for ADHD were excluded from the study sample	**		**	51 (M = 38)	32 (M = 2)	OTUs: 740 in ADHD, 645 in NT Alpha Diversity Index ACE: no differences Chao1: no differences Shannon: no differences Simpson: no differences Correlations with Faecalibacterium Total CPDS score: R = −0.294 (p = 0.037) Hyperactivity: R = −0.564 (p < 0.001)
Zhai (39)	ASD diagnosis according to DSM- IV-TR and ICD-10 criteria	No reported symptoms of ASD or other neurological disorders	Stool samples: collected by parents and guardians in a small cooler (−4°C) provided by the researchers, and returned on the same day to laboratory where it was homogenized, divided into aliquots and stored at −80°C	Meta-genomic DNA was isolated using a FastDNA Spin Kit for Soil	QIIME software (version 1.9.1)	Subjects were excluded if they were taking mineral supplements or antibiotic medications a month prior to sampling, or if they were in treatment with probiotics and/or prebiotics Controls were excluded if they had stomach/gut problems (e.g. chronic diarrhea, constipation), or if they were related to an individual with autism (e.g., sibling, parent)	*	*	*	78 (M = 56)	58 (M = 31)	Alpha diversity: Chao1 index Increased taxa richness in ASD (p = 0.015) Shannon index:: Significantly higher microbial diversity in ASD (p < 0.001) At a genus level, ASD children had a significant increase in Bacteroides, Parabacteroides, Sutterella, Lachnospira, Bacillus, Biophila, Lactococcus, Lachnobacterium, and Oscillospira (p < 0.01)
Liu (38)	ASD diagnosis according to DSM-5 and ICD-10 criteria	Neuro-typical (NT): typically developing children, without an autism diagnosis and not directly related to an autistic individual	Stool sample: collected and transported to the laboratory for processing within 30 min, where 200 mg samples were preserved in fecal bacteria DNA storage tubes and stored at −80°C	Microbial DNA was extracted from 200 mg fecal samples using the QIAamp Fast DNA Stool Mini Kit	UPARSE	Subjects were excluded if they had a history of use of nutritional supplements or were under special diets None of the included subjects was treated with antibiotics, antifungals, probiotics or prebiotics for at least 3 months before sampling	**	*	*	30 (M = 25)	20 (M = 16)	Alpha diversity SOBS, Chao and ACE indexes showed no significant differences Beta diversity Different overall composition At philum level in ASD Firmicutes were decreased (p < 0.05) while Acidobacteria were increased (p < 0.05) At family level no difference in Bacteroidaceae At taxa level in ASDVeillonellaceae and Enterobacteriaceae were increased (p < 0.05), while Ruminococcaceae, Streptococcaceae, Peptostreptococcaceae and Erysipelotrichaceae were decreased (p < 0.05)
Pulikkan (36)	ASD diagnosis according to DSM-5 criteria using CARS, AIIMS- modified INDT-ASD, and ISAA	Healthy siblings or blood relatives to the ASD children	Stool sample collected from each individual after morning breakfast and were stored at −80°C within 2 h of collection until further processing	DNA from fecal samples extracted using QIAamp Stool Mini Kit	QIIME	All ASD children had normal omnivore native diet similar to healthy subjects and were not on gluten-free diet (GFD) None of the participants took any antibiotic, anti- inflammatory, or antioxidant medication for 1 month prior to the sample collection	**		*	30 (M = 28)	24 (M = 15)	Alpha Diversity Index Shannon: no differences Observed: no differences Phylogenetic diversity: no differences Beta Diversity Principal Component AnalysisPC1 (38.1%): no differences PC2 (11.08%): no differences PC3 (7.43%): ASD (p = 0.023)Key families bacteria in differentiating ASD and healthy samples were Prevotellaceae, Lactobacillaceae, Mogibacillaceae
Zhang (37)	ASD diagnosis according to the DSM-5 criteria	NT children with no major psychiatric condition according to medical examination and parent interview recruited from 2 kindergartens	Stool samples collected at home by parents and immediately deep frozen, shipped to the laboratory on the same day and stored at −80°C until DNA extraction	Not specified	QIIME's RDP Classifier	None of the included subjects took antibiotics, antipsychotics, probiotics nor prebiotics in the past month prior to sample collection. Children with coeliac disease special diet (such as ketogenic diet) were excluded	***		**	35 (M = 28)	6 (M = 28)	Alpha diversity Shannon index revealed no significant differences between ASD and control group Beta diversity Bacterial microbiota of ASD clusters apart from control group (p = 0.02) At level of phylum, the ratioBacteroidetes/Firmicutes was higher in ASD (p ≤ 0.05), due to an increased relative abundance of Bacteroidetes (p ≤ 0.05)
Luna (35)	ASD diagnosis based on the Autism Diagnostic Observation Schedule, and a diagnosis of FGID	Neuro-typical (NT) control recruited at the outpatients pediatric GI suite, subdivided according to presence of FGID	Biopsy specimens: placed, immediately after collection, in 2 mL of saline on ice and transported to a laboratory for processing within 15 min, and stored at −80°C	Tissue specimens were processed through MO BIO Power Soil	Modified version of the UPARSE algorithm	None of the participants was taking antibiotics, steroids, nor had any GI infection during the 3 months prior to sample collection	**	*	*	14 (M = 14)	Total = 21 (M = 18) 15 (NT with FGID) 6 (NT without FGID)	Beta Diversity (performed by PCA) Increased in ASD with FGID Clostridium lituseburense (p = 0.002) Lachnoclostridium bolteae (p = 0.017) Lachnoclostridium hathewayi(p = 0.03) Clostridium aldenense (p = 0.38) Flavonifractor plautii (p = 0.038) Terrisporobacter (p = 0.045), after removing a single subjects in NT- FGIDDecreased in ASD with FGID Dorea formicigenerans (p = 0.006) Blautia luti (p = 0.025)Sutterella (p = 0.025)
Son (34)	ASD probands from Simons Simplex Collection diagnosed via Autism Diagnostic Observational Schedule, and subdivided according to the presence of FGID	Neuro-typical (NT) siblings recruited via Simons Simplex Collection registry through the Interactive Autism Network, and subdivided according to the presence of FGID	Stool samples: 2 ml of stool collected in a stool “hat” placed in the toilet and immediately transferred into a vial that contained 10 ml of RNA later for metagenomics studies Samples were shipped in cold packs overnight to the laboratory	Fecal DNA was extracted from stool samples immediately upon arrival using ZR Fecal DNA MiniPrep	UCHIME	All included subjects were off probiotics and antibiotics for at least 1 month	**		*	Total = 59 (M = 51):25 (ASD with FGID)34 (ASD without FGID)	Total = 37 (M = 16) 13 (NT with FGID) 31 (NT without FGID)	Alpha Diversity Index Chao1: no differences Shannon: no differences Beta Diversity PERMANOVA: No differences Binomial regression analysis: No differences Exploratory analysis: No differences
ORAL MICROBIOMA
Hicks (40)	ASD defined by clinician consensus using the DSM-5 criteria	Children with negative ASD screening based on the Modified Checklist for Autism in Toddlers-Revised, and children who met typical developmental milestones on standardized physician assessment	Saliva samples: collected at the time of enrollment, following an oral water rinse, using an ORAcollect swab from the sublingual and parotid regions of the mouth in a non-fasting state. Swabs were stored at −20°C prior to processing	Salivary RNA was extracted using a standard Trizol technique and the RNeasy mini column	RNA reads	Children with feeding tube dependence, active tooth	*		*	180 (M = 153)	106 (M = 64) healthy subjects	Alpha diversity No difference in Shannon index at both species and phylum level Beta diversity Bray-Curtis index showed significant differences between the ASD, TD and DD groups (p = 0.04) 12 taxa were different between the ASD, TD and DD groups (FDR < 0.05); 3 taxa were different between the ASD group and the DD group (FDR≤0.05); only Planctomycetes differed between ASD and both the TD (FDR = 0.001) and DD group (FDR = 0.02) No differences observed in the Firmicutes/Bacteroides ratio
Qiao (41)	ASD diagnosis according to the DSM-5 criteria, confirmed with the ICD-10 criteria	Healthy children recruited from primary schools	Saliva samples: 1 ml of non-stimulated, naturally outflowed saliva collected and transferred into 1.5 ml sterile tubes Dental plaques samples: first permanent molars isolated with cotton rolls and gentle air-drying Supra-gingival plaques obtained separately from caries-free molars in 4 quadrants per subject with sterile Gracey curettes, and pooled All samples were immediately placed on ice, transported to the laboratory within 2 h, and stored at −80°C	DNA from dental and salivary samples was extracted with the OMEGA-soil DNA Kit	QIIME (version 1.9.1)	Cases were excluded if in treatment with antibiotics within 3 months before the study, if treated with local antimicrobial agents within 2 weeks, or if they had been using any medication for ASD, sedatives, or were under gluten-free/casein-free (GF/CF) diet or were using probiotics Controls were excluded if they had any systemic or local disorders that caused oral mucosal lesions (e.g. lichen planus), if they had periodontal pockets ≥4 mm, acute oral infection (e.g. abscess), evidence of oral candidiasis, if they were in receiving antibiotics within 3 months before the study, or were treated with local antimicrobial agents within 2 weeks	***	*	*	32 (M = 27)	27 (M = 21)	Alpha Diversity Index dental ACE: ASD < control (p < 0.05) Shannon: ASD < control (p < 0.05) Shannoneven: ASD < control (p < 0.05) Alpha Diversity Index saliva ACE: no differences Shannon: no differences Shannoneven: no differences Beta Diversity Index in ASD Streptococcus and Haemophilus (high levels) Prevotella, Selenomonas, Actinomyces, Porphyromonas, and Fusobacterium (low levels) Rothia: high levels in dental, low levels in saliva

^§Newcastle-Ottawa Quality Assessment scale for Cohort and Case-Control Studies (NOS). Studies can be assigned a maximum of 4 stars in the Selection section, 2 stars in the Comparability section, and 3 stars in either the Outcome or the Exposure section.