Understanding the functional implications of genes' variants in autism heterogeneity is challenging. Gene set analysis examines the cumulative effect of multiple functionally converging genes. Here we explored whether a multi-step analysis could identify gene sets with different loads of protein-altering variants (PAVs) between two subgroups of autistic children. After subdividing our sample (n = 71, 3–12 years) based on higher (> 80; n = 43) and lower ( $\leqslant$ 80; n = 28) intelligence quotient (IQ), a gene set variant enrichment analysis identified gene sets with significantly different incidence of PAVs between the two subgroups of autistic children. Significant gene sets were then clustered into modules of genes. Their brain expression was investigated according to the BrainSpan Atlas of the Developing Human Brain. Next, we extended each module by selecting the genes that were spatio-temporally co-expressed in the developing brain and physically interacting with those in modules. Last, we explored the incidence of autism susceptibility genes within original and extended modules. Our analysis identified 38 significant gene sets (FDR, q < 0.05). They clustered in four modules involved in ion cell communication, neurocognition, gastrointestinal function, and immune system. Those modules were highly expressed in specific brain structures across development. Spatio-temporal brain co-expression and physical interactions identified extended genes' clusters with over-represented autism susceptibility genes. Overall, our unbiased approach identified modules of genes functionally relevant to autism pathophysiology, possibly implicating them in phenotypic variability across subgroups. The findings also suggest that autism diversity likely originates from multiple interacting pathways. Future research could leverage this approach to identify genetic pathways relevant to autism subtyping.

Summary

This study investigates how genetic differences contribute to the diversity of autism by examining gene variants in two groups of autistic children with different IQ levels.
By using an unbiased approach, we found 38 gene sets with significant differences in possibly damaging variants between the subgroups.
These gene sets were grouped into four key modules related to brain cell communication, neurocognition, gut function, and the immune system.
The modules were strongly expressed in certain brain regions and stages of development, and interact with genes including known autism susceptibility genes.
This approach offers valuable insights into how genetic factors may lead to different autism subtypes and could guide future research into autism's underlying biological mechanisms.

1 Introduction

Autism is a neurodevelopmental condition characterized by relevant heterogeneity across multiple levels of analysis. Its phenotypic presentation can drastically differ depending on the degree of both core and co-occurring features, such as language abilities (Naigles et al. 2017), intellectual profile (Wolff et al. 2022), motor coordination difficulties (Craig et al. 2021; Wang, Petrulla, et al. 2022), and adaptive functioning (Mandelli et al. 2023) (with this respect, see the LIMA approach, (Mandelli et al. 2024)). Critical heterogeneity has been found at deeper levels of investigation, including neurophysiology (Geschwind and Levitt 2007; Di Martino et al. 2014) and genetics (An and Claudianos 2016; Havdahl et al. 2021). Along with those aspects, it is also important to consider the dimension of chronogeneity, that is, the variation in autism manifestation in relation to the dimension of time and, markedly, of development (Georgiades et al. 2017). Despite this multi-level heterogeneity, the role of genetic factors in autism etiology is unquestioned. The heritability of the condition is high, with estimates ranging between 64% and 91% (Steffenburg et al. 1989; Le Couteur et al. 1996; Taniai et al. 2008; Lichtenstein et al. 2010; Hallmayer et al. 2011; Nordenbæk et al. 2014; Colvert et al. 2015; De Rubeis et al. 2014). Over recent years, hundreds of de novo and rare inherited variants have been identified as major contributors to individual autism risk (Iossifov et al. 2014; Sanders et al. 2015; Yuen et al. 2017; Ruzzo et al. 2019; Doan et al. 2019; Satterstrom et al. 2020; Choi and An 2021). The expression of those autism-associated genes has been reported to be preeminent at prenatal and early postnatal stages, substantiating the early onset of the condition (Willsey et al. 2013; Courchesne et al. 2019). However, although with smaller effect in comparison to that of de novo and rare inherited variants, several studies have demonstrated that heritability in autism is largely due to common variants, that is, to the genetic variations that are commonly present in the general population (Gaugler et al. 2014; de la Torre-Ubieta et al. 2016; Grove et al. 2019). Thus, as for other neurodevelopmental conditions, there is considerable evidence indicating the high heritability of autism and its polygenic nature, with both rare and common variants having a critical role (Cirnigliaro et al. 2023).

In the last 10 years, a body of research has leveraged genetic heterogeneity to parse autism phenotypic presentations (Jeste and Geschwind 2014; Chawner et al. 2021; Warrier et al. 2022; Di Giovanni et al. 2023; Bertelsen et al. 2021). Nevertheless, given the heterogeneity of the condition, the identification of the functional implications of genes related to autism requires large-scale studies, mostly involving thousands of participants.

An alternative perspective is shifting the focus from single genes to networks of genes that converge in functionally relevant biological processes, for example, by using gene set analysis. Gene sets are groups of genes that share a common biological function and are pre-defined based on prior biological knowledge (Subramanian et al. 2005). The gene set approach permits examination of the combined effect of multiple DNA variants that cumulatively impact biological pathways potentially relevant to autism (An and Claudianos 2016).

In this study, we explored the hypothesis of whether a multi-step unbiased analysis could identify gene sets relevant to autism subtyping and functionally characterize them in terms of biological processes and their spatio-temporal co-expression in the brain. Since we intended to evaluate the gene sets' cumulative effects, we considered all the protein-altering variants (PAVs), regardless of both their frequency in the population and their effect on the proteins.

The present work is split into different objectives. First, after dividing our clinical sample of autistic children into two subgroups based on their standardized measures of intelligence quotient (IQ), we identified gene sets that presented a different load of PAVs between participants with higher IQ and lower IQ, using a data-driven enrichment analysis. With this regard, previous research has reported significant associations between genetic variations in autism and cognitive heterogeneity (Wang et al. 2013; Iossifov et al. 2014). The obtained gene sets were hierarchically clustered into modules based on their PAVs enrichment. We then characterized each of those modules with a label representative of the biological processes that best characterized them.

Next, to explore the functional implications of the modules, we assessed their expression profiles in the brain according to the BrainSpan Atlas of the Developing Human Brain (n.d.). More specifically, we determined whether genes in the modules were expressed above expectations in brain structures through different developmental stages, to gain insights about their potential involvement in brain development.

Afterwards, to explore more extensively the functional interplay between the modules, we extended each module by selecting those genes which both physically interact at the protein level and show a highly correlated spatio-temporal brain expression profile with those in the modules. Among the extended set of genes, we assessed their spatial and temporal co-expression in the brain according to the BrainSpan Atlas of the Developing Human Brain and their physical interaction at the protein level according to the bioGRID database (Stark et al. 2006). Understanding how genes relate to each other during neurodevelopment is essential to identify what are potential shared biological processes (Mahfouz et al. 2015) and how they could interact along their functional pathways to impact the final outcome. Genes that are co-expressed in the brain are indeed more likely to act together in converging processes. In this sense, the evidence of the corresponding proteins' interactions represents a confirmation of this functional association (Oldham et al. 2008).

Lastly, we investigated whether the genes in the modules and those in the extended set of their co-expressed interactors were enriched in candidate genes highly associated with autism susceptibility according to the Simons Foundation Autism Research Initiative (SFARI) database (n.d.), which constantly integrates genetic information from multiple research studies.

2 Methods

We summarized our methods in a workflow diagram (Figure 1).

Details are in the caption following the image — **FIGURE 1**
Open in figure viewer PowerPoint

Workflow diagram depicting the study's methods.

2.1 Participants

A total of 71 autistic children aged 3–12 years were involved in this cross-sectional study. Participants were consecutively recruited at the Child Psychopathology Unit of Scientific Institute, IRCCS Eugenio Medea (Bosisio Parini, Italy), over a 36-month period between July 2016 and July 2019, as a part of a larger study (Crippa et al. 2021). Autistic children were admitted to inpatient/outpatient units of our institute either for assessment or for rehabilitation programs. All participants had been previously diagnosed at our institute on the basis of a consensus “best estimate” DSM-5 clinical diagnostic process informed by, but not dependent on, scores on the Autism Diagnostic Observation Schedule-Second Edition (ADOS-2) (Lord et al. 2012). Participants were excluded in case a well-defined genetic disorder was detected. Further exclusion criteria were the use of medication affecting the central nervous system, the presence of significant sensory impairment (e.g., blindness, deafness), abnormalities detected by MRI, and suffering from chronic or acute medical illness. All participants were drug-naïve.

This research was conducted in accordance with the ethical standards of the 1964 Declaration of Helsinki and later amendments and was approved by the Ethics Committee of our Institute “Comitato Etico IRCCS E. Medea—Sezione Scientifica Associazione La Nostra Famiglia” (Prot. N.33/18—CE). Informed written consent was collected by all of the participants' parents or legal guardians before participation.

2.2 Measures

All children underwent an assessment of the IQ level. The specific assessment tool was chosen according to the participants' developmental level among the Griffiths Mental Development Scales (Griffiths 1970), Wechsler Preschool and Primary Scale of Intelligence (WPPSI-III) (Wechsler 2002), and Wechsler Intelligence Scale for Children-IV (WISC-IV) (Wechsler 2012). Autism core symptoms severity was rated using the calibrated severity scores for the total, the social affect, and the restricted and repetitive behavior domains of the ADOS-2 (Lord et al. 2012), and the Social Responsiveness Scales (SRS) (Constantino and Gruber 2005). In the context of a broader research project (Crippa et al. 2021), the participants' motor abilities were also evaluated using the Movement Assessment Battery for Children-2 (MABC2) (Henderson et al. 2007), the sensorimotor subtests of the Developmental Neuropsychological Assessment Second Edition (NEPSY-II) (Korkman et al. 2007), the Beery-Buktenica Developmental Test of Visual-Motor Integration (VMI) (Beery and Beery 2004), and the Developmental Coordination Disorder Questionnaire (DCDQ) (Wilson et al. 2014), a motor screener filled by caregivers. Lastly, information about the familial socioeconomic status (SES) was collected using the Hollingshead scale for parental employment (Hollingshead 1975). Table S1 provides a description of the sample size for each of the investigated features.

2.3 Biological Samples

Blood or saliva samples (7 and 64, respectively) were obtained from participants for DNA extraction, which was performed in-house at the Molecular Biology Laboratory of our Institute. Blood was collected in tubes with EDTA. Genomic DNA was extracted from blood samples through the GenElute Blood Genomic DNA kit (Sigma). DNA from each sample was eluted in 50 μL of 10 mM Tris (pH 9.0)/0.5 mM EDTA and stored at −20°C. The Oragene OG-500 kit (DNA Genotek) was used to collect saliva samples. Genomic DNA was extracted from saliva samples by precipitation in ethanol following the manufacturer's protocol, resuspended in 50 μL of 10 mM Tris (pH 9.0)/1 mM EDTA and stored at −20°C.

2.4 Variants Identification

Samples' DNA were screened by using a targeted next generation sequencing approach with a targeted design that enables analysis of only the disease-associated targets, the SureSelect Focused Exome, with approximately 5200 disease-associated genes (see Table S2 for a complete list of genes). The sequencing libraries were prepared from genomic DNA by using a Sure Select enrichment system (Agilent Technologies). Targeted libraries were run on the NextSeq platform according to the manufacturer's instructions (Illumina, San Diego, CA, USA). The sequenced reads were then aligned to reference target regions, and variants were called with the BWA enrichment application (which include also GATK for variant calling) available on BaseSpace Onsite (Illumina, San Diego, CA, USA). Erroneous variant calls were discarded. Furthermore, we did not consider the Y chromosome because some variations were retrieved also for women.

Variants' annotation was performed by ANNOVAR (Wang et al. 2010) with a refGene database generated by converting the human reference genome (GRCh37/hg19) GFF3 and FASTA files to ANNOVAR file format using the gff3toGenePred package (Grüning et al. 2018). PAVs were then selected as those occurring in the panel exons and annotated as indels, stop codon gain, stop codon loss, or non-synonymous. Lastly, we quantified variants for each participant, controlling for the presence of rare variants (i.e., not reported or with a frequency below 0.01 in gnomAD, [Karczewski et al. 2020]) or pathogenic variants (annotated as Pathogenic or Likely Pathogenic according to ACMG-AMP guidelines, obtained from the InterVar database) (Li and Wang 2017).

2.5 Gene Sets Definition

We obtained 29,227 gene sets corresponding to ontology terms or pathways and their associated genes. To this purpose, we interrogated multiple sources, namely: Gene Ontology (Ashburner et al. 2000) (version 2019-03), HPO Phenotype (Kanehisa et al. 2010) (version 2019-03), Kegg Pathway Database (Kanehisa et al. 2010) (2019-2-25) and Reactome Pathway Database (Fabregat et al. 2018) (Physical Entity Identifier mapping files, updated to 2019-03).

2.6 Evaluation of Gene Sets Variants Abundance in a Group of Subjects

To keep into account variants zygosity, we calculated the number of variants V for a given gene set S and a group of subjects G as follows:

V=\frac{\left({\sum}_{i\in G}{\sum}_{g\in {S}_A}{V}_{ig}\right)\left({\sum}_{i\in G}{\sum}_{g\in {S}_A}2{L}_g\right)+\left({\sum}_{i\in G}{\sum}_{g\in {S}_X}{V}_{ig}\right)\left({\sum}_{i\in G}{\sum}_{g\in {S}_A}{N}_i{L}_g\right)}{\sum_{i\in G}{\sum}_{g\in {S}_A}2{L}_g+{\sum}_{i\in G}{\sum}_{g\in {S}_A}{N}_i{L}_g}

(1)

where V_ig is the number of alternative alleles observed for subject i in gene g, L_g is the length of gene g, and N_i is 2 if the subject i is a female or 1 otherwise. Summations are over the subjects in the group G and the genes over the gene-set S (S_A for genes in autosomal chromosomes and S_X for genes on the X chromosome). The same procedure was used to evaluate the number of variants in a gene set for a single subject (in this case G only contains one subject). We define a gene set enrichment score for a group or a single subject as the number of variants in the gene set calculated following Equation (1) divided by the total number of variants for the group or for that specific subject.

2.7 Gene Set Variants Enrichment Analysis

The goal of the gene set variants enrichment analysis (GSVEA) is to determine whether the abundance of variants observed in a given gene set differs in two subject groups. To assess whether the variants abundance in a gene set S is significantly different between two groups of participants G1 and G2, a Fisher's exact test is applied to the contingency table defined as:

	Number of variants in the gene set	Number of variants not in the gene set
G1	V_G1	V_A1–V_G1
G2	V_G2	V_A2–V_G2

V_G1 and V_G2 are calculated according to Equation (1) for the gene set S while V_A1 and V_A2 are calculated according to Equation (1) but for the gene set comprising all the genes in the exome panel. When multiple gene sets are tested, false discovery rate (FDR) is applied to the Fisher test p values and q values are reported.

2.8 Gene Sets and Subjects Clustering

Hierarchical clustering has been consistently performed in this study using the Ward algorithm (Murtagh and Legendre 2014) applied to the euclidean distance. Optimal ranks have been determined by silhouette analysis (Rousseeuw 1987).

2.9 Gene Expression in the Brain During Neurodevelopmental Stages

We used the ABAEnrichment R package (version 1.20.0) (Grote et al. 2016) to characterize the brain expression of the genes included in the modules according to the BrainSpan Atlas of the Developing Human Brain (Chawner et al. 2021). BrainSpan Atlas includes five developmental stages, namely prenatal, infant (0–2 years), child (3–11 years), adolescent (12–19 years) and adult (> 19 years). According to the study's aims, we limited our analysis to the four developmental stages.

2.10 Genes Spatio-Temporal Co-Expression in the Developing Brain

For each gene in the exome panel, we built an expression vector extracting from the BrainSpan Atlas of the Developing Human Brain the expression levels measured in each available structure for the first four developmental stages. For any pair of genes in a set, we estimated their co-expression in time and space by calculating the Pearson correlation coefficient (p) between their expression vectors. Correlation p values are also computed and corrected for multiple testing (FDR). Gene pairs with p > 0.7 and q values < 0.05 are considered to be co-expressed.

2.11 Genes Protein–Protein Interactions

The bioGRID database (Stark et al. 2006) was used to identify whether couples of genes had physical protein–protein interactions. Genes having at least one physical interaction with the genes in the modules were considered as interactors in the present study.

2.12 Incidence of Autism-Related Genes

Lastly, we analyzed the incidence of genes with a gene score of 1 (High Confidence category) in the SFARI database (https://sfari.org/ accessed July 9, 2024) in our gene modules and their extended networks. Fisher exact tests were performed considering all the genes listed in the exome panel and, in case of multiple testing, p values were corrected according to Bonferroni.

3 Results

3.1 Clinical and Motor Differences Between Subgroups of Participants

Table 1 summarizes the sociodemographic, clinical, and motor characteristics of the total sample and for the two subgroups of participants with lower (≤ 80, n = 28) and higher IQ (> 80, n = 43), respectively. The two subgroups were balanced in autism severity as assessed with both the ADOS-2 calibrated severity scores and the SRS, and in the male to female ratio. Participants with higher IQ had significantly higher scores on SES (p = 0.035) and were slightly older than subjects with lower IQ (p = 0.057). Demographically, the entire sample was Caucasian.

TABLE 1. Sample description.

	Autism, TOT (n = 71)	Autism, IQ ≤ 80 (n = 28)	Autism, IQ > 80 (n = 43)	Subgroup comparisons
Age	7.6 (2.4) [3.2–11.9]	6.9 (2.6) [3.2–11.9]	8.1 (2.1) [3.8–11.6]	0.057^c
Sex (M:F)	61:10	25:3	36:7	0.510^c
IQ	86.1 (21.8) [47–134]	64.7 (9.6) [47–80]	100.0 (15.0) [81–134]	< 0.001 ^c
SES	58.5 (18.1) [20–90]	52.7 (16.9) [20–90]	62.5 (18.1) [30–90]	0.035 ^c
Autism core characteristics
ADOS-2 total CSS	6.2 (1.7) [3–10]	6.1 (1.7) [4–10]	6.2 (1.7) [3–9]	0.505^c
ADOS-2 SA CSS	6.4 (1.9) [3–10]	6.5 (1.9) [3–10]	6.4 (1.9) [3–10]	0.696^c
ADOS-2 RRB CSS	6.3 (2.1) [1–10]	5.9 (2.7) [1–10]	6.5 (1.6) [1–9]	0.756^c
SRS^a	73.8 (29.8) [26–148]	68.11 (31.1) [26–148]	77.6 (28.7) [26–142]	0.200^c
Motor skills
DCDQ—Control^a	17.5 (5.2) [7–30]	18.1 (5.2) [7–28]	17.1 (5.2) [8–30]	0.566^d
DCDQ—Fine motor/handwriting^a	12.2 (4.6) [4–20]	11.1 (5.3) [4–19]	12.9 (3.9) [4–20]	0.261^d
DCDQ—General coordination^a	14.5 (4.6) [7–24]	15.4 (5.1) [7–24]	13.9 (4.2) [9–23]	0.378^d
MABC 2—Manual dexterity^b	6.0 (2.8) [1–13]	4.5 (2.2) [2–10]	6.8 (2.7) [1–13]	0.003 ^d
MABC 2—Aiming and catching^b	7.1 (2.6) [2–13]	7.3 (2.6) [3–13]	6.9 (2.7) [2–13]	0.691^d
MABC 2—Balance^b	6.3 (2.3) [1–14]	6.0 (1.9) [1–10]	6.5 (2.5) [1–14]	0.566^d
NEPSY II—Visuomotor precision^b	6.8 (2.3) [1–14]	6.1 (1.8) [1–10]	7.2 (2.4) [2–14]	0.122^d
NEPSY II—Fingertip tapping^b	11.4 (2.8) [1–15]	10.3 (3.5) [2–15]	12 (2.3) [1–15]	0.159^d
NEPSY II—Manual motor sequences^b	9.2 (3.4) [1–15]	8.0 (3.3) [3–14]	10 (3.2) [2–15]	0.086^d
NEPSY II—Imitating hand positions^b	6.8 (4.7) [1–15]	3.9 (3.9) [1–12]	8.3 (4.4) [1–15]	0.002 ^d
VMI total^b	93.8 (17.8) [57–148]	85.4 (13.5) [66–114]	98.6 (18.3) [57–148]	0.012 ^d

Note: Data are expressed as Mean (SD) [range]. Mann–Whitney U test or t-test was used for quantitative variables, chi-squared test for the categorical variable sex. ^araw score; ^bstandard score; ^cp value; ^dq value, false discovery rate corrected. Bold font: p values < 0.05.
Abbreviations: ADOS: autism diagnostic observation schedule; CSS: calibrated severity scores; DCDQ: Developmental Coordination Disorder Questionnaire; MABC2: Movement Assessment Battery for Children 2; NEPSY-II: Developmental Neuropsychological Assessment Second Edition; RRB: restricted repetitive behaviors; SA: social affect; SES: socio-economic status; SRS: Social Responsiveness Scales; VMI: Beery-Buktenica Developmental Test of Visual-Motor Integration.

With respect to the motor skills, participants with higher IQ presented higher scores on MABC2 Manual Dexterity, NEPSY-II Imitating Hand Positions, and VMI (all FDR q < 0.05).

3.2 Gene Sets Showing Differential PAVs Enrichment Between Autism Subgroups Are Involved in Ion Cell Communication, Neurocognition, Gastrointestinal Function, and Immune System

In order to check for the presence of rare or pathogenic variants in the two subgroups of participants, we quantified variants for each subject, applying progressively stricter selection criteria. When considering all variants (Figure S1, Panel A), participants with higher IQ showed a slightly greater number of variants. This non-significant difference progressively decreased when we focused on rare variants (Figure S1, Panel B) or pathogenic variants (Figure S1, Panel C). When we filtered for both frequency and pathogenicity (Figure S1, Panel D), the trend reversed, with slightly, although not significantly, more variants observed in the low IQ group. The application of this last criterion selected 144 variants reported in Table S3.

After this preliminary check, we performed a gene set variants enrichment analysis (GSVEA). In brief, the goal of this analysis was to determine whether the abundance of PAVs observed in a given gene set differed in the two subgroups of autistic participants. We performed the analysis for nearly 30.000 functionally characterized gene sets obtained from public databases of ontologies and pathways (see methods). Among the considered gene sets, 38 showed a significantly different incidence of PAVs between the two subgroups of autistic participants (FDR, q < 0.05). For each subject, we calculated the proportion of PAVs—enrichment score—in each significant gene set. We then applied an optimal hierarchical clustering procedure to the resulting matrix at both gene sets and subjects level (Figure 2A). The significant gene sets resulted grouped in five clusters (terms' clusters, TC). Two of them—TC2 and TC3—were merged in a single cluster, given their high intersection level (Figure 2B), resulting in a total of four modules, which were operationally defined as the union of the corresponding significantly enriched gene sets. These four modules grouped together genes involved in a variety of processes and functions related to ion cell communication, neurocognition, gastrointestinal function, and immune system, respectively, and were thus labeled accordingly. The data-driven identification of these four modules was based on the initial top-down splitting of our sample in two subgroups on the basis of their standardized measures of IQ. With a bottom-up approach, we also explored whether the modules' enrichment effectively identified clusters of participants with phenotypic differences. At the subject level, participants were clustered in three groups (subjects' clusters, SC). These SCs had a non-random distribution of subjects with IQ > 80 (Fisher exact test; p < 0.001), which resulted particularly represented in SC2 (Figure 2C). A logistic regression analysis based on the five TCs enrichment scores was also performed. With a leave-one-out cross validation approach, the regression model showed an accuracy of 0.704 (area under the receiver operating characteristic curve, ROC AUC = 0.74) in predicting IQ group membership (Figure 2D). Participants in the three SCs did not differ significantly in ADOS total or subscales scores (Kruskal–Wallis test; p > 0.05).

3.3 Genes in the Modules Are Expressed in the Brain in the Developmental Period

We then aimed to understand whether the genes included in the four modules could be relevantly expressed in brain regions throughout development. To this end, we leveraged the spatial gene expression data provided by the BrainSpan Atlas of the Developing Human Brain to link our gene modules to brain structures across development. Given the childhood-onset nature of the condition, we specifically considered the first four developmental stages listed in the BrainSpan Atlas: prenatal, infancy (0–2 years), childhood (3–11 years), adolescence (12–19 years).

Results of the ABAEnrichment analysis revealed that genes in most of the modules were significantly more expressed across specific brain structures and developmental stages, except for the immune system module. An overview of the significantly enriched brain structures in each developmental stage can be found in Table 2. Genes in the ion cell communication module appear to be particularly expressed in the prenatal period and in adolescence across different brain structures. Genes in the neurocognition module are mainly expressed in the hippocampus and in various primary and associative cortical areas across infancy and childhood. Genes of the gastrointestinal module are extensively expressed from infancy to adolescence in the cerebellum, subcortical nuclei, and cortical structures. Last, genes included in the immune system module are not expressed higher than expected in either any brain structure or developmental stage.

TABLE 2. Gene modules' expression enrichment across brain structures and developmental stages.

	Ion cell communication				Neurocognition				Immune system				Gastro-intestinal
	Pre-natal	Infant 0–2 years	Child 3–11 years	Adol 12–19 years	Pre-natal	Infant 0–2 years	Child 3–11 years	Adol 12–19 years	Pre-natal	Infant 0–2 years	Child 3–11 years	Adol 12–19 years	Pre-natal	Infant 0–2 years	Child 3–11 years	Adol 12–19 years
CBC_cerebellar cortex														0.032 1.138	0.028 1.156	0.043 1.177
MD_mediodorsal nucleus of thalamus		0.01 3.25	0.031 2.884	0.037 2.78
STR_striatum	0.01 1.476	0.021 3.06													0.037 1.302
AMY_amygdaloid complex														0.042 1.23	0.035 1.216
HIP_hippocampus (hippocampal formation)						0.016 2.36	0.037 2.552							0.018 1.246	0.005 1.268	0.023 1.454
M1C_primary motor cortex (area M1, area 4)	0.03 1.389			0.016 2.98		0.025 2.001								0.04 1.249	0.013 1.465
S1C_primary somatosensory cortex (area S1, areas 3, 1, 2)				0.013 2.997		0.022 2.265								0.041 1.237
V1C_primary visual cortex (striate cortex, area V1/17)			0.029 2.945	0.013 3.015											0.042 1.185
A1C_primary auditory cortex (core)	0.004 1.56													0.049 1.2	0.024 1.215
OFC_orbital frontal cortex				0.037 2.78		0.017 2.331	0.049 2.524
VFC_ventrolateral prefrontal cortex				0.034 2.863		0.019 2.01	0.012 1.706							0 1.468
DFC_dorsolateral prefrontal cortex				0.04 2.775	0.038 1.614		0.008 3.292
MFC_anterior (rostral) cingulate (medial prefrontal) cortex							0.008 1.75								0.004 1.307	0.008 1.531
ITC_inferolateral temporal cortex (area TEv, area 20)	0.047 1.437			0.035 2.821		0.01 2.398
STC_posterior (caudal) superior temporal cortex (area 22c)						0.018 2.321	0.023 1.666
IPC_posteroventral (inferior) parietal cortex	0.004 1.56					0.008 2.408									0.025 1.201

Note: Sampled brain structures are listed on the left, and investigated age ranges are reported in columns. p values (bold text) and fold values (plain text) are reported for brain structures and age ranges where a significant enrichment for that gene module was found.

3.4 Identification of Extended Modules Through Physical Protein–Protein Interactions and Brain Spatio-Temporal Co-Expression

Next, we aimed at identifying the genes which are both interacting with those in the four modules and showing a high level of co-expression in the developing brain. To this purpose, we used the protein–protein interaction database bioGRID, and the expression data from the BrainSpan Atlas of the Developing Human Brain. For each gene in a module, we selected its direct physical interactors which also presented a significant level of spatio-temporal co-expression in the brain with that gene (see Section 2). Each module was then extended with its co-expressed interactors (Table 3, see Table S4 for a complete list of genes).

TABLE 3. Number of genes in each module and in each module extended with its co-expressed interactors.

	Number of genes in each module	Number of genes in each extended module
Ion cell communication	36	70
Neurocognition	20	149
Gastrointestinal function	94	331
Immune system	80	207
Total	219	590

We then collected all the genes in the four extended modules (n = 590) and built a co-expression matrix by calculating the spatio-temporal co-expression for each pair (Figure 3A). Hierarchical clustering identified three genes' clusters (GC) among this extended set of genes. Genes in clusters GC1 and GC2 showed high within-cluster and low between-cluster co-expression levels, while those in cluster GC3 showed a general intermediate level of co-expression both internally and with the other two clusters. Figure 3B depicts the annotated protein–protein interaction network for the three clusters, displaying high connection levels for clusters GC1 and GC2.

3.5 SFARI Genes Are Overrepresented in the Extended Networks of Genes in the Modules

Looking at the incidence of autism-related genes, nine out of the 219 genes in the modules (4.11%) were also annotated in the SFARI database in the high confidence category. This proportion was higher than expected in a random sample of genes, but did not reach the level of significance (OR = 1.534; p = 0.2063). When considering the four extended modules, high-confidence SFARI genes were significantly overrepresented (32 out of 590, 5.42%; OR = 2.296, p = 0.0001; see Table S5 for a complete list of genes).

Looking at the distribution of these SFARI genes in the three co-expression clusters, they were significantly overrepresented in cluster GC1 (28 out of 362, 7.73%; OR = 3.40; corrected p < 0.001) while not in GC2 (1 out of 153, 0.65%) and GC3 (3 out of 63, 4.76%). Moreover, within GC1, the enrichment of SFARI genes was significant in the extended networks of the neurocognition module (10 out of 128, 7.81%; OR = 3.114; Bonferroni corrected p = 0.033) and gastrointestinal module (16 out of 246, 6.50%; OR = 2.615; Bonferroni corrected p = 0.014) but not in the immune system and ion cell communication modules.

4 Discussion

In this work, we identified four modules of genes with different PAVs' load in two autism subgroups with higher and lower IQ levels by using an unsupervised approach. Importantly, the identified modules grouped together genes involved in a variety of biological processes—ion cell communication, neurocognition, gastrointestinal function, and immune system—that have been previously reported to be atypical in autism. Here, for the first time to our knowledge, we provide preliminary evidence of the possible implication of these biological processes in the differential phenotypic manifestations of two discrete autism subgroups.

The first gene module—ion cell communication—includes genes involved in ion homeostasis, transport, and signaling (e.g., ATP1A2 and ATP1A3—subunits of a sodium-potassium pump, CACNA1C—subunit of a calcium voltage-gated channel). Those genes have been extensively associated with autism (Schmunk and Gargus 2013; Daghsni et al. 2018; Evangelho et al. 2023), mostly for their crucial role in synaptic functioning (De Rubeis et al. 2014; Bourgeron 2015; Satterstrom et al. 2020). In the very early developmental stages, genes within this module encode proteins with major contributions to neural proliferation, migration, and differentiation (Smith and Walsh 2020), including voltage-gated ion channels that regulate the propagation of action potentials and pacemaking, also in relation to cardiac membrane excitability. With this respect, an association between autism and congenital heart disease has been reported (Gu et al. 2023; Sigmon et al. 2019; Splawski et al. 2004) and functionally confirmed based on the evidence of convergent genetic pathways (De Rubeis et al. 2014; Rosenthal et al. 2021). Genes in this module were significantly enriched in many brain regions during the prenatal stage according to our brain expression analysis, confirming their early involvement in development. Brain expression results also identified a significant enrichment of those genes during adolescence. With this respect, an association between the expression of ion-related genes in adolescence and neuropsychiatric conditions has been observed (Clifton et al. 2021). Interestingly, our results also align with previous research indicating an association between variants in ion channels-related genes and both higher IQ and lower levels of repetitive behaviors in autism (Lee et al. 2021), confirming the relevance of this gene module in distinguishing different phenotypic manifestations of the condition.

The neurocognition module encompasses gene sets related to various alterations in cognitive functions, including lack of insight, anomia, agnosia, alexia, as well as some core autistic features such as circumscribed behaviors and collectionism. Coherently with their involvement in cognitive functions, we found that the brain expression of these genes is mainly localized in cortical regions. Interestingly, many of the included genes are involved in the causation of different forms of neurodegenerative dementia (e.g., TREM2, GRN, PSEN1, C9ORF72, MAPT), including frontotemporal dementia (Firdaus and Li 2024; Hinz and Geschwind 2017; DeJesus-Hernandez et al. 2011). Common etiopathological mechanisms have been hypothesized between neurodegenerative dementia and neurodevelopmental conditions, including autism (Li et al. 2022; Nadeem et al. 2021; Fumagalli and Crippa 2022), based on the observations of partially overlapping symptoms (Nadeem et al. 2021; Rhodus et al. 2020) and increased occurrence of dementia in autistic adults (Vivanti et al. 2021). Moreover, these genes are also involved in biological processes specific to neurodevelopment, such as the production of neurotrophic factors (Wang, Chen, et al. 2022), synaptic development and refinement (Tian et al. 2024; Filipello et al. 2018), and broad synaptic functioning (Saura et al. 2011), which is coherent with their potential involvement in autism.

Third, the finding of a gene module related to gastrointestinal function is in line with the relatively recent hypothesis of gut involvement in the etiological mechanisms of autism (Dargenio et al. 2023; Bjørklund et al. 2020). Gastrointestinal symptoms are frequent in a large percentage of autistic individuals (Chaidez et al. 2014) and are associated with the severity of the condition (Adams et al. 2011). Alterations in gut microbiome are also commonly reported in autism (Xu et al. 2019; Liu et al. 2019). The impact of the gut microbiota on the brain, mainly via the bidirectional microbiota-gut-brain axis, has been broadly investigated in neurodevelopmental conditions (Morais et al. 2021). Our expression analyses revealed that genes in the gastrointestinal module were extensively expressed in the brain across all developmental stages. The present result, therefore, extends the previous finding of a surprisingly high proportion of autism-associated genes expressed both in the brain and in the gastrointestinal tract (Niesler and Rappold 2021), further indicating a common genetic pathway for alterations in these two systems.

Lastly, the immune system has been extensively involved in autism, both as a candidate etiological mechanism, for example, through maternal immune activation, and as a potential endophenotype, with converging evidence of immune dysregulation in autistic individuals (Meltzer and Van de Water 2017; Onore et al. 2012; Ormstad et al. 2018; Lombardo et al. 2018; Careaga et al. 2017; Masi et al. 2017). Moreover, genes controlling innate and adaptive immunity have been previously associated with autism (Horiuchi et al. 2021; Arenella et al. 2023), but also with autistic traits in the general population (Arenella et al. 2022). Interestingly, some of the immune genes possibly involved in autism are also implicated in neurodevelopmental processes such as neuronal plasticity, with their strongest expression in the brain during gestational and early postnatal age (Arenella et al. 2023; Estes and McAllister 2015). In contrast with those findings, we did not find here genes within the immune module to be particularly expressed in the brain along the neurodevelopmental stages. With this respect, it is important to have in mind that the gene expression profiles considered in this work are those reported in the BrainSpan Atlas of the Developing Human Brain for typically developing individuals. It is therefore possible that the presence of genetic variants in autistic subjects contributes to change the typical gene expression patterns, leading to the immune genes' over-expression in the brain that is reported elsewhere in the literature (Garbett et al. 2008; Voineagu et al. 2011). Furthermore, this result could also suggest a systemic rather than brain-localized role of the immune system in autism. Indeed, immune dysregulation can impact brain function through different pathways, including the gut-brain axis where the immune system is a crucial mediator (Azhari et al. 2019; Powell et al. 2017). Moreover, some circulating cytokines can reach the brain and inhibit neurogenesis or promote neuron death, whereas endogenous anti-brain antibodies may be produced, altering the development or function of neurons (Meltzer and Van de Water 2017; Estes and McAllister 2015).

After having identified the four modules with a data-driven analysis based on the initial top-down splitting of our sample on IQ, we applied a bottom-up approach to explore whether the variants enrichment could effectively identify different clusters of participants. Coherently with our initial stratification, we found that participants with IQ > 80 were particularly represented in one of the three SC, as also confirmed by a predictive logistic regression model. While this result aligns with our criterion for subgrouping participants, the moderate accuracy of the regression model suggests that other factors are implicated in the between-groups differences associated with the modules' enrichment. Nevertheless, the SC were not highly differentiated in terms of autism symptom severity. Future extensions of this work should combine different units of analysis (e.g., cells, circuits/networks, neurobiology) to further clarify linkages between genotype and clinical phenotypes.

In addition to pinpointing the gene modules and discussing their relationship to phenotypic characteristics of autism, we thoroughly explored the functional interplay between the modules by considering an extended set of genes including the co-expressed interactors of the genes in each module. We discovered that most of the genes within this extended set presented high levels of spatio-temporal brain co-expression across development. Consistently, genes with higher co-expression correlation levels also highly interacted at the level of protein–protein interactions. Likewise, genes from different modules were found to be mutually interacting. Complementing earlier reports (De Rubeis et al. 2014; Lombardo et al. 2017; Gupta et al. 2014), this work therefore provides further evidence supporting the idea that the biology of autism could likely implicate altered interactions between different modules rather than the existence of independent disrupted pathways, which cumulatively contribute to the clinical outcome.

Lastly, we found that SFARI high-confidence autism-related genes were overrepresented in the total set of genes in the extended modules, but not in the restricted group of genes in the modules. This suggests a significant connection, but no intersection, between the four pinpointed modules and genes previously associated with autism susceptibility. Nonetheless, this result also indicates that our unbiased method could effectively identify sets of genes in close relationship with autism-related genes, uncovering their potential role in shaping the heterogeneous autistic phenotype. Importantly, the distribution of SFARI high-confidence genes within the total set of genes in the four extended modules was not stochastic. Indeed, based on their spatio-temporal co-expression levels, we found three different gene clusters: one with higher interaction levels but relatively few SFARI genes, a second one with fewer interactions and fewer SFARI genes, and a third cluster displaying high levels of interaction and overrepresented SFARI genes. In this last cluster, the autism-related genes were specifically overrepresented within the extended networks of the neurocognition and gastrointestinal genes. Interestingly, the interaction between SFARI and neurocognition-related genes is in line with a previous large-scale exome sequencing study in autism (De Rubeis et al. 2014), which found a significant protein–protein interaction between autism genes and MAPT, a gene implicated in neurodegenerative disorders, that is also part of our neurocognition module. Previous work has also highlighted that above 90% of the 62 highest-ranking autism risk genes in the SFARI database are expressed in both brain and gastrointestinal tissues (Niesler and Rappold 2021), substantiating their interactions.

5 Limitations

Some limitations of the present study should be considered. The main limitation is related to the small sample sizes of participant groups; the present findings, therefore, must be considered preliminary. Future extensions of this work should include additional cohorts from publicly available sequencing cohorts (e.g., SPARK) as replication datasets. Nonetheless, this work serves as a “proof-of-concept” demonstrating the potential value of our analytical method, even within a restricted yet phenotypically well-characterized group of autistic participants. With respect to the genetic data, our analysis was limited to a subsection of the whole exome (~5200 genes) screened with a panel that exclusively enables analysis of disease-associated target genes. Notwithstanding this limitation, the present preliminary findings align with previous evidence from whole-exome sequencing studies. However, it should be noted that whole-exome/genome analysis could have yielded an even more comprehensive pattern of results. Moreover, by only considering the load of PAVs in a gene set, we neglect possible differences between each variant effect on the gene set associated function. The availability of this information could have facilitated functional connections between genetics and phenotypic manifestations. Lastly, at the present stage, the analysis of differences in the load of PAVs between subgroups of participants relies solely on IQ scores. Future extensions of this work should use the present multi-step approach to investigate genetic differences also between autism subtypes defined by their core and non-core features in motor, language, intellectual, and adaptive functioning.

6 Conclusions

In the present study, we have shown evidence that an unbiased, multi-step analysis could identify sets of genes involved in neurocognition, ion cell communication, gastrointestinal function, and immune system that are potentially related to the phenotypic differences among autistic children with different IQ levels. Besides being early expressed in the brain, such biological pathways are spatio-temporally co-expressed and highly interconnected with each other and with many autism-related genes. These findings therefore support the hypothesis that the diversity in autism likely originates from multiple interacting pathways that could be altered at various levels rather than deriving from a series of independent functional cascades with cumulative impact on the final outcome. Although these observations should be considered with caution, future research could leverage the present approach to identify genetic pathways relevant to autism subtyping, investigating a link between genetic profiles and distinct biotypes of autistic individuals.

Author Contributions

Conceptualization: L.F., U.P., and A.C. Methodology: G.S., L.F., M.Ma., R.G., M.V., U.P., and A.C. Software: L.F., M.Ma., and U.P. Formal analysis: G.S., L.F., M.Ma., and U.P. Investigation: R.G., M.V., S.B.C., L.V., E.M., M.N., M.Mo., and A.C. Data curation: G.S., L.F., S.B.C., U.P., and A.C. Writing – original draft preparation: G.S., U.P., and A.C. Writing – review and editing: G.S., L.F., U.P., and A.C. Visualization: G.S., L.F., U.P., and A.C. Supervision: U.P. and A.C. Project administration: U.P. and A.C. Funding acquisition: A.C.

Conflicts of Interest

The authors declare no conflicts of interest.

Open Research

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Supporting Information

Filename	Description
aur70086-sup-0001-FigureS1.pdfPDF document, 50.7 KB	Figure S1. Number of variants in the two subgroups of participants according to their frequency and pathogenicity.
aur70086-sup-0002-TableS1.xlsxExcel 2007 spreadsheet , 16.3 KB	Table S1. Description of the sample size for each of the investigated clinical and motor features.
aur70086-sup-0003-TableS2.xlsxExcel 2007 spreadsheet , 69.4 KB	Table S2. Panel of genes under analysis in the present work.
aur70086-sup-0004-TableS3.xlsxExcel 2007 spreadsheet , 20.4 KB	Table S3. Description of the 144 PAVs found in our sample after filtering for both frequency and pathogenicity.
aur70086-sup-0005-TableS4.xlsxExcel 2007 spreadsheet , 40.5 KB	Table S4. Genes in the modules and their co-expressed interactors (extended modules).
aur70086-sup-0006-TableS5.xlsxExcel 2007 spreadsheet , 20.4 KB	Table S5. Genes in the modules and their SFARI interactors.

Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

References

Adams, J. B., L. J. Johansen, L. D. Powell, D. Quig, and R. A. Rubin. 2011. “Gastrointestinal Flora and Gastrointestinal Status in Children With Autism–Comparisons to Typical Children and Correlation With Autism Severity.” BMC Gastroenterology 11: 1–13.
10.1186/1471-230X-11-22
PubMed Web of Science® Google Scholar
An, J. Y., and C. Claudianos. 2016. “Genetic Heterogeneity in Autism: From Single Gene to a Pathway Perspective.” Neuroscience & Biobehavioral Reviews 68: 442–453.
10.1016/j.neubiorev.2016.06.013
CAS PubMed Web of Science® Google Scholar
Arenella, M., G. Cadby, W. De Witte, et al. 2022. “Potential Role for Immune-Related Genes in Autism Spectrum Disorders: Evidence From Genome-Wide Association Meta-Analysis of Autistic Traits.” Autism 26, no. 2: 361–372.
10.1177/13623613211019547
PubMed Web of Science® Google Scholar
Arenella, M., R. Matuleviciute, R. Tamouza, et al. 2023. “Immunogenetics of Autism Spectrum Disorder: A Systematic Literature Review.” Brain, Behavior, and Immunity 114: 488–499.
10.1016/j.bbi.2023.09.010
CAS PubMed Web of Science® Google Scholar
Ashburner, M., C. A. Ball, J. A. Blake, et al. 2000. “Gene Ontology: Tool for the Unification of Biology.” Nature Genetics 25, no. 1: 25–29.
10.1038/75556
CAS PubMed Web of Science® Google Scholar
Azhari, A., F. Azizan, and G. Esposito. 2019. “A Systematic Review of Gut-Immune-Brain Mechanisms in Autism Spectrum Disorder.” Developmental Psychobiology 61, no. 5: 752–771.
10.1002/dev.21803
PubMed Web of Science® Google Scholar
Beery, K. E., and N. A. Beery. 2004. The Beery-Buktenica Developmental Test of Visual-Motor Integration: Administration, Scoring, and Teaching Manual. 5th ed. Modern Curriculum Press.
Google Scholar
Bertelsen, N., I. Landi, R. A. Bethlehem, et al. 2021. “Imbalanced Social-Communicative and Restricted Repetitive Behavior Subtypes of Autism Spectrum Disorder Exhibit Different Neural Circuitry.” Communications Biology 4, no. 1: 574.
10.1038/s42003-021-02015-2
CAS PubMed Web of Science® Google Scholar
Bjørklund, G., L. Pivina, M. Dadar, et al. 2020. “Gastrointestinal Alterations in Autism Spectrum Disorder: What Do We Know?” Neuroscience & Biobehavioral Reviews 118: 111–120.
10.1016/j.neubiorev.2020.06.033
PubMed Web of Science® Google Scholar
“BrainSpan Atlas of the Developing Human Brain.” n.d. https://www.brainspan.org.
Google Scholar
Bourgeron, T. 2015. “From the Genetic Architecture to Synaptic Plasticity in Autism Spectrum Disorder.” Nature Reviews Neuroscience 16, no. 9: 551–563.
10.1038/nrn3992
CAS PubMed Web of Science® Google Scholar
Careaga, M., S. Rogers, R. L. Hansen, D. G. Amaral, J. Van de Water, and P. Ashwood. 2017. “Immune Endophenotypes in Children With Autism Spectrum Disorder.” Biological Psychiatry 81, no. 5: 434–441.
10.1016/j.biopsych.2015.08.036
CAS PubMed Web of Science® Google Scholar
Chaidez, V., R. L. Hansen, and I. Hertz-Picciotto. 2014. “Gastrointestinal Problems in Children With Autism, Developmental Delays or Typical Development.” Journal of Autism and Developmental Disorders 44: 1117–1127.
10.1007/s10803-013-1973-x
PubMed Web of Science® Google Scholar
Chawner, S. J., J. L. Doherty, R. J. Anney, et al. 2021. “A Genetics-First Approach to Dissecting the Heterogeneity of Autism: Phenotypic Comparison of Autism Risk Copy Number Variants.” American Journal of Psychiatry 178, no. 1: 77–86.
10.1176/appi.ajp.2020.20010015
PubMed Web of Science® Google Scholar
Choi, L., and J. An. 2021. “Genetic Architecture of Autism Spectrum Disorder: Lessons From Large-Scale Genomic Studies.” Neuroscience & Biobehavioral Reviews 128: 244–257.
10.1016/j.neubiorev.2021.06.028
CAS PubMed Web of Science® Google Scholar
Cirnigliaro, M., T. S. Chang, S. A. Arteaga, et al. 2023. “The Contributions of Rare Inherited and Polygenic Risk to ASD in Multiplex Families.” Proceedings of the National Academy of Sciences 120, no. 31: e2215632120.
10.1073/pnas.2215632120
CAS PubMed Web of Science® Google Scholar
Clifton, N. E., L. Collado-Torres, E. E. Burke, et al. 2021. “Developmental Profile of Psychiatric Risk Associated With Voltage-Gated Cation Channel Activity.” Biological Psychiatry 90, no. 6: 399–408.
10.1016/j.biopsych.2021.03.009
CAS PubMed Web of Science® Google Scholar
Colvert, E., B. Tick, F. McEwen, et al. 2015. “Heritability of Autism Spectrum Disorder in a UK Population-Based Twin Sample.” JAMA Psychiatry 72: 415.
10.1001/jamapsychiatry.2014.3028
PubMed Web of Science® Google Scholar
Constantino, J. N., and C. P. Gruber. 2005. Social Responsiveness Scale (SRS) Manual. Western Psychological Services.
Google Scholar
Courchesne, E., T. Pramparo, V. H. Gazestani, M. V. Lombardo, K. Pierce, and N. E. Lewis. 2019. “The ASD Living Biology: From Cell Proliferation to Clinical Phenotype.” Molecular Psychiatry 24, no. 1: 88–107.
10.1038/s41380-018-0056-y
PubMed Web of Science® Google Scholar
Craig, F., A. Crippa, M. Ruggiero, et al. 2021. “Characterization of Autism Spectrum Disorder (ASD) Subtypes Based on the Relationship Between Motor Skills and Social Communication Abilities.” Human Movement Science 77: 102802.
10.1016/j.humov.2021.102802
PubMed Web of Science® Google Scholar
Crippa, A., F. Craig, S. Busti Ceccarelli, et al. 2021. “A Multimethod Approach to Assessing Motor Skills in Boys and Girls With Autism Spectrum Disorder.” Autism 25, no. 5: 1481–1491.
10.1177/1362361321995634
PubMed Web of Science® Google Scholar
Daghsni, M., M. Rima, Z. Fajloun, et al. 2018. “Autism Throughout Genetics: Perusal of the Implication of Ion Channels.” Brain and Behavior: A Cognitive Neuroscience Perspective 8, no. 8: e00978.
10.1002/brb3.978
Google Scholar
Dargenio, V. N., C. Dargenio, S. Castellaneta, et al. 2023. “Intestinal Barrier Dysfunction and Microbiota–Gut–Brain Axis: Possible Implications in the Pathogenesis and Treatment of Autism Spectrum Disorder.” Nutrients 15, no. 7: 1620.
10.3390/nu15071620
CAS PubMed Web of Science® Google Scholar
de la Torre-Ubieta, L., H. Won, J. L. Stein, and D. H. Geschwind. 2016. “Advancing the Understanding of Autism Disease Mechanisms Through Genetics.” Nature Medicine 22, no. 4: 345–361.
10.1038/nm.4071
CAS PubMed Web of Science® Google Scholar
De Rubeis, S., X. He, A. P. Goldberg, et al. 2014. “Synaptic, Transcriptional and Chromatin Genes Disrupted in Autism.” Nature 515, no. 7526: 209–215.
10.1038/nature13772
CAS PubMed Web of Science® Google Scholar
DeJesus-Hernandez, M., I. R. Mackenzie, B. F. Boeve, et al. 2011. “Expanded GGGGCC Hexanucleotide Repeat in Noncoding Region of C9ORF72 Causes Chromosome 9p-Linked FTD and ALS.” Neuron 72, no. 2: 245–256.
10.1016/j.neuron.2011.09.011
CAS PubMed Web of Science® Google Scholar
Di Giovanni, D., R. Enea, V. Di Micco, A. Benvenuto, P. Curatolo, and G. L. Emberti. 2023. “Using Machine Learning to Explore Shared Genetic Pathways and Possible Endophenotypes in Autism Spectrum Disorder.” Genes 14, no. 2: 313.
10.3390/genes14020313
CAS PubMed Web of Science® Google Scholar
Di Martino, A., C. Yan, Q. Li, et al. 2014. “The Autism Brain Imaging Data Exchange: Towards a Large-Scale Evaluation of the Intrinsic Brain Architecture in Autism.” Molecular Psychiatry 19, no. 6: 659–667.
10.1038/mp.2013.78
CAS PubMed Web of Science® Google Scholar
Doan, R. N., E. T. Lim, S. De Rubeis, et al. 2019. “Recessive Gene Disruptions in Autism Spectrum Disorder.” Nature Genetics 51, no. 7: 1092–1098.
10.1038/s41588-019-0433-8
CAS PubMed Web of Science® Google Scholar
Estes, M. L., and A. K. McAllister. 2015. “Immune Mediators in the Brain and Peripheral Tissues in Autism Spectrum Disorder.” Nature Reviews Neuroscience 16, no. 8: 469–486.
10.1038/nrn3978
CAS PubMed Web of Science® Google Scholar
Evangelho, V. G. O., M. L. Bello, H. C. Castro, and M. R. Amorim. 2023. “Gene Set Enrichment Analysis Indicates Convergence in the mTOR Signalling Pathway Between Syndromic and Non-Syndromic Autism.” Neuroscience Informatics 3, no. 2: 100119.
10.1016/j.neuri.2023.100119
Google Scholar
Fabregat, A., F. Korninger, G. Viteri, et al. 2018. “Reactome Graph Database: Efficient Access to Complex Pathway Data.” PLoS Computational Biology 14, no. 1: e1005968.
10.1371/journal.pcbi.1005968
PubMed Web of Science® Google Scholar
Filipello, F., R. Morini, I. Corradini, et al. 2018. “The Microglial Innate Immune Receptor TREM2 Is Required for Synapse Elimination and Normal Brain Connectivity.” Immunity 48, no. 5: 979–991.
10.1016/j.immuni.2018.04.016
CAS PubMed Web of Science® Google Scholar
Firdaus, Z., and X. Li. 2024. “Unraveling the Genetic Landscape of Neurological Disorders: Insights Into Pathogenesis, Techniques for Variant Identification, and Therapeutic Approaches.” International Journal of Molecular Sciences 25, no. 4: 2320.
10.3390/ijms25042320
CAS PubMed Web of Science® Google Scholar
Fumagalli, G. G., and A. Crippa. 2022. “The Search for a Link Between Autism Spectrum Disorder and Neurodegenerative Dementia.” Minerva Psychiatry 63, no. 3: 189–196.
10.23736/S2724-6612.22.02370-3
Google Scholar
Garbett, K., P. J. Ebert, A. Mitchell, et al. 2008. “Immune Transcriptome Alterations in the Temporal Cortex of Subjects With Autism.” Neurobiology of Disease 30, no. 3: 303–311.
10.1016/j.nbd.2008.01.012
CAS PubMed Web of Science® Google Scholar
Gaugler, T., L. Klei, S. J. Sanders, et al. 2014. “Most Genetic Risk for Autism Resides With Common Variation.” Nature Genetics 46, no. 8: 881–885.
10.1038/ng.3039
CAS PubMed Web of Science® Google Scholar
Georgiades, S., S. L. Bishop, and T. Frazier. 2017. “Editorial Perspective: Longitudinal Research in Autism–Introducing the Concept of ‘Chronogeneity’.” Journal of Child Psychology and Psychiatry 58, no. 5: 634–636.
10.1111/jcpp.12690
PubMed Web of Science® Google Scholar
Geschwind, D. H., and P. Levitt. 2007. “Autism Spectrum Disorders: Developmental Disconnection Syndromes.” Current Opinion in Neurobiology 17, no. 1: 103–111.
10.1016/j.conb.2007.01.009
CAS PubMed Web of Science® Google Scholar
Griffiths, R. 1970. “The Abilities of Young Children.” A Study in Mental Measurement.
Google Scholar
Grote, S., K. Prüfer, J. Kelso, and M. Dannemann. 2016. “ABAEnrichment: An R Package to Test for Gene Set Expression Enrichment in the Adult and Developing Human Brain.” Bioinformatics 32, no. 20: 3201–3203.
10.1093/bioinformatics/btw392
CAS PubMed Web of Science® Google Scholar
Grove, J., S. Ripke, T. D. Als, et al. 2019. “Identification of Common Genetic Risk Variants for Autism Spectrum Disorder.” Nature Genetics 51, no. 3: 431–444.
10.1038/s41588-019-0344-8
CAS PubMed Web of Science® Google Scholar
Grüning, B., R. Dale, A. Sjödin, et al. 2018. “Bioconda: Sustainable and Comprehensive Software Distribution for the Life Sciences.” Nature Methods 15, no. 7: 475–476.
10.1038/s41592-018-0046-7
CAS PubMed Web of Science® Google Scholar
Gu, S., A. Katyal, Q. Zhang, W. Chung, S. Franciosi, and S. Sanatani. 2023. “The Association Between Congenital Heart Disease and Autism Spectrum Disorder: A Systematic Review and Meta-Analysis.” Pediatric Cardiology 44, no. 5: 1092–1107.
10.1007/s00246-023-03146-5
PubMed Web of Science® Google Scholar
Gupta, S., S. E. Ellis, F. N. Ashar, et al. 2014. “Transcriptome Analysis Reveals Dysregulation of Innate Immune Response Genes and Neuronal Activity-Dependent Genes in Autism.” Nature Communications 5, no. 1: 5748.
10.1038/ncomms6748
CAS PubMed Web of Science® Google Scholar
Hallmayer, J., S. Cleveland, A. Torres, et al. 2011. “Genetic Heritability and Shared Environmental Factors Among Twin Pairs With Autism.” Archives of General Psychiatry 68: 1095–1102.
10.1001/archgenpsychiatry.2011.76
PubMed Web of Science® Google Scholar
Havdahl, A., M. Niarchou, A. Starnawska, M. Uddin, C. van der Merwe, and V. Warrier. 2021. “Genetic Contributions to Autism Spectrum Disorder.” Psychological Medicine 51, no. 13: 2260–2273.
10.1017/S0033291721000192
CAS PubMed Web of Science® Google Scholar
Henderson, S. E., D. A. Sugden, and A. L. Barnett. 2007. Movement Assessment Battery for Children. 2nd ed. Pearson.
Google Scholar
Hinz, F. I., and D. H. Geschwind. 2017. “Molecular Genetics of Neurodegenerative Dementias.” Cold Spring Harbor Perspectives in Biology 9, no. 4: a023705.
10.1101/cshperspect.a023705
PubMed Web of Science® Google Scholar
Hollingshead, A. B. 1975. Four Factor Index of Social Status. Yale University.
10.1016/0148-9062(75)90418-0
Google Scholar
Horiuchi, F., Y. Yoshino, H. Kumon, et al. 2021. “Identification of Aberrant Innate and Adaptive Immunity Based on Changes in Global Gene Expression in the Blood of Adults With Autism Spectrum Disorder.” Journal of Neuroinflammation 18: 1–12.
10.1186/s12974-021-02154-7
PubMed Web of Science® Google Scholar
Iossifov, I., B. J. O'roak, S. J. Sanders, et al. 2014. “The Contribution of de Novo Coding Mutations to Autism Spectrum Disorder.” Nature 515, no. 7526: 216–221.
10.1038/nature13908
CAS PubMed Web of Science® Google Scholar
Jeste, S. S., and D. H. Geschwind. 2014. “Disentangling the Heterogeneity of Autism Spectrum Disorder Through Genetic Findings.” Nature Reviews Neurology 10, no. 2: 74–81.
10.1038/nrneurol.2013.278
PubMed Web of Science® Google Scholar
Kanehisa, M., S. Goto, M. Furumichi, M. Tanabe, and M. Hirakawa. 2010. “KEGG for Representation and Analysis of Molecular Networks Involving Diseases and Drugs.” Nucleic Acids Research 38, no. suppl_1: D355–D360.
10.1093/nar/gkp896
CAS PubMed Google Scholar
Karczewski, K. J., L. C. Francioli, G. Tiao, et al. 2020. “The Mutational Constraint Spectrum Quantified From Variation in 141,456 Humans.” Nature 581: 434–443.
10.1038/s41586-020-2308-7
CAS PubMed Web of Science® Google Scholar
Korkman, M., U. Kirk, and S. Kemp. 2007. NEPSY-II. 2nd ed. Harcourt Assessment.
Google Scholar
Le Couteur, A., A. Bailey, S. Goode, et al. 1996. “A Broader Phenotype of Autism: The Clinical Spectrum in Twins.” Journal of ChildPsychology and Psychiatry 37: 785–801.
10.1111/j.1469-7610.1996.tb01475.x
CAS Google Scholar
Lee, J., S. Ha, J. Ahn, S. Lee, J. R. Choi, and K. Cheon. 2021. “The Role of Ion Channel-Related Genes in Autism Spectrum Disorder: A Study Using Next-Generation Sequencing.” Frontiers in Genetics 12: 595934.
10.3389/fgene.2021.595934
CAS PubMed Web of Science® Google Scholar
Li, C., D. Pang, J. Lin, T. Yang, and H. Shang. 2022. “Shared Genetic Links Between Frontotemporal Dementia and Psychiatric Disorders.” BMC Medicine 20, no. 1: 131.
10.1186/s12916-022-02335-y
PubMed Web of Science® Google Scholar
Li, Q., and K. Wang. 2017. “InterVar: Clinical Interpretation of Genetic Variants by ACMG-AMP 2015 Guideline.” American Journal of Human Genetics 100: 1–14.
10.1016/j.ajhg.2017.01.004
Google Scholar
Lichtenstein, P., E. Carlström, M. Råstam, C. Gillberg, and H. Anckarsäter. 2010. “The Genetics of Autism Spectrum Disorders and Related Neuropsychiatric Disorders in Childhood.” American Journal of Psychiatry 167: 1357–1363.
10.1176/appi.ajp.2010.10020223
PubMed Web of Science® Google Scholar
Liu, F., J. Li, F. Wu, H. Zheng, Q. Peng, and H. Zhou. 2019. “Altered Composition and Function of Intestinal Microbiota in Autism Spectrum Disorders: A Systematic Review.” Translational Psychiatry 9, no. 1: 43.
10.1038/s41398-019-0389-6
PubMed Web of Science® Google Scholar
Lombardo, M. V., E. Courchesne, N. E. Lewis, and T. Pramparo. 2017. “Hierarchical Cortical Transcriptome Disorganization in Autism.” Molecular Autism 8: 1–17.
10.1186/s13229-017-0147-7
PubMed Web of Science® Google Scholar
Lombardo, M. V., H. M. Moon, J. Su, T. D. Palmer, E. Courchesne, and T. Pramparo. 2018. “Maternal Immune Activation Dysregulation of the Fetal Brain Transcriptome and Relevance to the Pathophysiology of Autism Spectrum Disorder.” Molecular Psychiatry 23, no. 4: 1001–1013.
10.1038/mp.2017.15
CAS PubMed Web of Science® Google Scholar
Lord, C., M. Rutter, S. Goode, et al. 2012. “Autism Diagnostic Observation Schedule.” Journal of Autism and Developmental Disorders.
PubMed Web of Science® Google Scholar
Mahfouz, A., M. N. Ziats, O. M. Rennert, B. P. Lelieveldt, and M. J. Reinders. 2015. “Shared Pathways Among Autism Candidate Genes Determined by Co-Expression Network Analysis of the Developing Human Brain Transcriptome.” Journal of Molecular Neuroscience 57, no. 4: 580–594.
10.1007/s12031-015-0641-3
CAS PubMed Web of Science® Google Scholar
Mandelli, V., I. Landi, E. M. Busuoli, E. Courchesne, K. Pierce, and M. V. Lombardo. 2023. “Prognostic Early Snapshot Stratification of Autism Based on Adaptive Functioning.” Nature Mental Health 1, no. 5: 327–336.
10.1038/s44220-023-00056-6
Web of Science® Google Scholar
Mandelli, V., I. Severino, L. Eyler, K. Pierce, E. Courchesne, and M. V. Lombardo. 2024. “A 3D Approach to Understanding Heterogeneity in Early Developing Autisms.” Molecular Autism 15, no. 1: 41.
10.1186/s13229-024-00613-5
PubMed Web of Science® Google Scholar
Masi, A., N. Glozier, R. Dale, and A. J. Guastella. 2017. “The Immune System, Cytokines, and Biomarkers in Autism Spectrum Disorder.” Neuroscience Bulletin 33: 194–204.
10.1007/s12264-017-0103-8
CAS PubMed Web of Science® Google Scholar
Meltzer, A., and J. Van de Water. 2017. “The Role of the Immune System in Autism Spectrum Disorder.” Neuropsychopharmacology 42, no. 1: 284–298.
10.1038/npp.2016.158
CAS PubMed Web of Science® Google Scholar
Morais, L. H., H. L. Schreiber IV, and S. K. Mazmanian. 2021. “The Gut Microbiota–Brain Axis in Behaviour and Brain Disorders.” Nature Reviews Microbiology 19, no. 4: 241–255.
10.1038/s41579-020-00460-0
CAS PubMed Web of Science® Google Scholar
Murtagh, F., and P. Legendre. 2014. “Ward's Hierarchical Agglomerative Clustering Method: Which Algorithms Implement Ward's Criterion?” Journal of Classification 31: 274–295.
10.1007/s00357-014-9161-z
Web of Science® Google Scholar
Nadeem, M. S., S. Hosawi, S. Alshehri, et al. 2021. “Symptomatic, Genetic, and Mechanistic Overlaps Between Autism and Alzheimer's Disease.” Biomolecules 11, no. 11: 1635.
10.3390/biom11111635
CAS PubMed Web of Science® Google Scholar
Naigles, L. R., R. Johnson, A. Mastergeorge, et al. 2017. “Neural Correlates of Language Variability in Preschool-Aged Boys With Autism Spectrum Disorder.” Autism Research 10, no. 6: 1107–1119.
10.1002/aur.1756
PubMed Web of Science® Google Scholar
Niesler, B., and G. A. Rappold. 2021. “Emerging Evidence for Gene Mutations Driving Both Brain and Gut Dysfunction in Autism Spectrum Disorder.” Molecular Psychiatry 26, no. 5: 1442–1444.
10.1038/s41380-020-0778-5
PubMed Web of Science® Google Scholar
Nordenbæk, C., M. Jørgensen, K. O. Kyvik, and N. Bilenberg. 2014. “A Danish Population-Based Twin Study on Autism Spectrum Disorders.” European Child and Adolescent Psychiatry 23: 35–43.
10.1007/s00787-013-0419-5
PubMed Web of Science® Google Scholar
Oldham, M. C., G. Konopka, K. Iwamoto, et al. 2008. “Functional Organization of the Transcriptome in Human Brain.” Nature Neuroscience 11, no. 11: 1271–1282.
10.1038/nn.2207
CAS PubMed Web of Science® Google Scholar
Onore, C., M. Careaga, and P. Ashwood. 2012. “The Role of Immune Dysfunction in the Pathophysiology of Autism.” Brain, Behavior, and Immunity 26, no. 3: 383–392.
10.1016/j.bbi.2011.08.007
CAS PubMed Web of Science® Google Scholar
Ormstad, H., V. Bryn, O. D. Saugstad, O. Skjeldal, and M. Maes. 2018. “Role of the Immune System in Autism Spectrum Disorders (ASD).” CNS & Neurological Disorders-Drug Targets (Formerly Current Drug Targets-CNS & Neurological Disorders) 17, no. 7: 489–495.
CAS PubMed Web of Science® Google Scholar
Powell, N., M. M. Walker, and N. J. Talley. 2017. “The Mucosal Immune System: Master Regulator of Bidirectional Gut–Brain Communications.” Nature Reviews Gastroenterology & Hepatology 14, no. 3: 143–159.
10.1038/nrgastro.2016.191
CAS PubMed Web of Science® Google Scholar
Rhodus, E. K., J. Barber, E. L. Abner, et al. 2020. “Behaviors Characteristic of Autism Spectrum Disorder in a Geriatric Cohort With Mild Cognitive Impairment or Early Dementia.” Alzheimer Disease & Associated Disorders 34, no. 1: 66–71.
10.1097/WAD.0000000000000345
PubMed Web of Science® Google Scholar
Rosenthal, S. B., H. R. Willsey, Y. Xu, et al. 2021. “A Convergent Molecular Network Underlying Autism and Congenital Heart Disease.” Cell Systems 12, no. 11: 1094–1107.
10.1016/j.cels.2021.07.009
CAS PubMed Web of Science® Google Scholar
Rousseeuw, P. J. 1987. “Silhouettes: A Graphical Aid to the Interpretation and Validation of Cluster Analysis.” Journal of Computational and Applied Mathematics 20: 53–65.
10.1016/0377-0427(87)90125-7
Web of Science® Google Scholar
Ruzzo, E. K., L. Pérez-Cano, J. Jung, et al. 2019. “Inherited and de Novo Genetic Risk for Autism Impacts Shared Networks.” Cell 178, no. 4: 850–866.
10.1016/j.cell.2019.07.015
CAS PubMed Web of Science® Google Scholar
Sanders, S. J., X. He, A. J. Willsey, et al. 2015. “Insights Into Autism Spectrum Disorder Genomic Architecture and Biology From 71 Risk Loci.” Neuron 87, no. 6: 1215–1233.
10.1016/j.neuron.2015.09.016
CAS PubMed Web of Science® Google Scholar
Satterstrom, F. K., J. A. Kosmicki, J. Wang, et al. 2020. “Large-Scale Exome Sequencing Study Implicates Both Developmental and Functional Changes in the Neurobiology of Autism.” Cell 180, no. 3: 568–584.
10.1016/j.cell.2019.12.036
CAS PubMed Web of Science® Google Scholar
Saura, C. A., E. Servián-Morilla, and F. G. Scholl. 2011. “Presenilin/γ-Secretase Regulates Neurexin Processing at Synapses.” PLoS One 6, no. 4: e19430.
10.1371/journal.pone.0019430
CAS PubMed Google Scholar
Schmunk, G., and J. J. Gargus. 2013. “Channelopathy Pathogenesis in Autism Spectrum Disorders.” Frontiers in Genetics 4: 222.
10.3389/fgene.2013.00222
PubMed Google Scholar
Sigmon, E. R., M. Kelleman, A. Susi, C. M. Nylund, and M. E. Oster. 2019. “Congenital Heart Disease and Autism: A Case-Control Study.” Pediatrics 144, no. 5: e20184114.
10.1542/peds.2018-4114
PubMed Web of Science® Google Scholar
“Simons Foundation Autism Research Initiative (SFARI) Database.” n.d. https://sfari.org.
Google Scholar
Smith, R. S., and C. A. Walsh. 2020. “Ion Channel Functions in Early Brain Development.” Trends in Neurosciences 43, no. 2: 103–114.
10.1016/j.tins.2019.12.004
CAS PubMed Web of Science® Google Scholar
Splawski, I., K. W. Timothy, L. M. Sharpe, et al. 2004. “CaV1. 2 Calcium Channel Dysfunction Causes a Multisystem Disorder Including Arrhythmia and Autism.” Cell 119, no. 1: 19–31.
10.1016/j.cell.2004.09.011
CAS PubMed Web of Science® Google Scholar
Stark, C., B. Breitkreutz, T. Reguly, L. Boucher, A. Breitkreutz, and M. Tyers. 2006. “BioGRID: A General Repository for Interaction Datasets.” Nucleic Acids Research 34, no. suppl_1: D535–D539.
10.1093/nar/gkj109
CAS PubMed Web of Science® Google Scholar
Steffenburg, S., C. Gillberg, L. Hellgren, et al. 1989. “A Twin Study of Autism in Denmark, Finland, Iceland, Norway and Sweden.” Journal of Child Psychology and Psychiatry 30: 405–416.
10.1111/j.1469-7610.1989.tb00254.x
CAS PubMed Web of Science® Google Scholar
Subramanian, A., P. Tamayo, V. K. Mootha, et al. 2005. “Gene Set Enrichment Analysis: A Knowledge-Based Approach for Interpreting Genome-Wide Expression Profiles.” Proceedings of the National Academy of Sciences 102, no. 43: 15545–15550.
10.1073/pnas.0506580102
CAS PubMed Web of Science® Google Scholar
Taniai, H., T. Nishiyama, T. Miyachi, M. Imaeda, and S. Sumi. 2008. “Genetic Influences on the Broad Spectrum of Autism: Study of Proband-Ascertained Twins.” American Journal ofMedical Genetics Part B-Neuropsychiatric Genetics 147B: 844–849.
10.1002/ajmg.b.30740
CAS PubMed Web of Science® Google Scholar
Tian, Y., X. Xiao, W. Liu, et al. 2024. “TREM2 Improves Microglia Function and Synaptic Development in Autism Spectrum Disorders by Regulating P38 MAPK Signaling Pathway.” Molecular Brain 17, no. 1: 12.
10.1186/s13041-024-01081-x
CAS PubMed Web of Science® Google Scholar
Vivanti, G., S. Tao, K. Lyall, D. L. Robins, and L. L. Shea. 2021. “The Prevalence and Incidence of Early-Onset Dementia Among Adults With Autism Spectrum Disorder.” Autism Research 14, no. 10: 2189–2199.
10.1002/aur.2590
PubMed Web of Science® Google Scholar
Voineagu, I., X. Wang, P. Johnston, et al. 2011. “Transcriptomic Analysis of Autistic Brain Reveals Convergent Molecular Pathology.” Nature 474, no. 7351: 380–384.
10.1038/nature10110
CAS PubMed Web of Science® Google Scholar
Wang, H. Z., H. Qin, W. Guo, J. Samuels, and Y. Y. Shugart. 2013. “New Insights Into the Genetic Mechanism of IQ in Autism Spectrum Disorders.” Frontiers in Genetics 4: 195.
10.3389/fgene.2013.00195
PubMed Google Scholar
Wang, K., M. Li, and H. Hakonarson. 2010. “ANNOVAR: Functional Annotation of Genetic Variants From High-Throughput Sequencing Data.” Nucleic Acids Research 38, no. 16: e164.
10.1093/nar/gkq603
CAS PubMed Web of Science® Google Scholar
Wang, L., J. Chen, Y. Hu, et al. 2022. “Progranulin Improves Neural Development via the PI3K/Akt/GSK-3β Pathway in the Cerebellum of a VPA-Induced Rat Model of ASD.” Translational Psychiatry 12, no. 1: 114.
10.1038/s41398-022-01875-4
CAS PubMed Google Scholar
Wang, L. A., V. Petrulla, C. J. Zampella, R. Waller, and R. T. Schultz. 2022. “Gross Motor Impairment and Its Relation to Social Skills in Autism Spectrum Disorder: A Systematic Review and Two Meta-Analyses.” Psychological Bulletin 148, no. 3–4: 273.
10.1037/bul0000358
PubMed Web of Science® Google Scholar
Warrier, V., X. Zhang, and P. Reed. 2022. “Genetic Correlates of Phenotypic Heterogeneity in Autism.” Nature Genetics 54, no. 9: 1293–1304.
10.1038/s41588-022-01072-5
CAS PubMed Web of Science® Google Scholar
Wechsler, D. 2002. Wechsler Preschool and Primary Scale of Intelligence (WPPSI- III). 3rd ed. Psychological Corporation.
Google Scholar
Wechsler, D. 2012. Wechsler Intelligence Scale for Children–IV (WISCIV) Italian Edition. Organizzazioni Speciali.
Google Scholar
Willsey, A. J., S. J. Sanders, M. Li, et al. 2013. “Coexpression Networks Implicate Human Midfetal Deep Cortical Projection Neurons in the Pathogenesis of Autism.” Cell 155, no. 5: 997–1007.
10.1016/j.cell.2013.10.020
CAS PubMed Web of Science® Google Scholar
Wilson, B. N., B. J. Kaplan, S. G. Crawford, and G. Roberts. 2014. “Cross-Cultural Adaptation, Reliability and Predictive Validity of the Italian Version of Developmental Coordination Disorder Questionnaire (DCDQ).” European Journal of Paediatric Neurology 18: 267–272.
10.1016/j.ejpn.2013.11.009
PubMed Web of Science® Google Scholar
Wolff, N., S. Stroth, I. Kamp-Becker, S. Roepke, and V. Roessner. 2022. “Autism Spectrum Disorder and IQ–A Complex Interplay, Front Psychiatry.” Frontiers in Psychiatry 13, no. 856084.
Google Scholar
Xu, M., X. Xu, J. Li, and F. Li. 2019. “Association Between Gut Microbiota and Autism Spectrum Disorder: A Systematic Review and Meta-Analysis.” Frontiers in Psychiatry 10: 473.
10.3389/fpsyt.2019.00473
PubMed Web of Science® Google Scholar
Yuen, R. K., D. Merico, M. Bookman, et al. 2017. “Whole Genome Sequencing Resource Identifies 18 New Candidate Genes for Autism Spectrum Disorder.” Nature Neuroscience 20, no. 4: 602–611.
10.1038/nn.4524
CAS PubMed Web of Science® Google Scholar

Early View

Online Version of Record before inclusion in an issue

Protein-Altering Variants' Analysis in Autism Subgroups Uncovers Early Brain-Expressed Gene Modules Relevant to Autism Pathophysiology

ABSTRACT

Summary

1 Introduction

2 Methods

2.1 Participants

2.2 Measures

2.3 Biological Samples

2.4 Variants Identification

2.5 Gene Sets Definition

2.6 Evaluation of Gene Sets Variants Abundance in a Group of Subjects

2.7 Gene Set Variants Enrichment Analysis

2.8 Gene Sets and Subjects Clustering

2.9 Gene Expression in the Brain During Neurodevelopmental Stages

2.10 Genes Spatio-Temporal Co-Expression in the Developing Brain

2.11 Genes Protein–Protein Interactions

2.12 Incidence of Autism-Related Genes

3 Results

3.1 Clinical and Motor Differences Between Subgroups of Participants

3.2 Gene Sets Showing Differential PAVs Enrichment Between Autism Subgroups Are Involved in Ion Cell Communication, Neurocognition, Gastrointestinal Function, and Immune System

3.3 Genes in the Modules Are Expressed in the Brain in the Developmental Period

3.4 Identification of Extended Modules Through Physical Protein–Protein Interactions and Brain Spatio-Temporal Co-Expression

3.5 SFARI Genes Are Overrepresented in the Extended Networks of Genes in the Modules

4 Discussion

5 Limitations

6 Conclusions

Author Contributions

Conflicts of Interest

Open Research

Data Availability Statement

Supporting Information

References

Figures

References

Information

About Wiley Online Library

Help & Support

Opportunities

Connect with Wiley