Multi-Omics Research in Autism

The Project

GEMMA, which stands for Genome, Environment, Microbiome and Metabolome in Autism, is a multicentre European Commission Project for exploring interactions between gut microbiome, metabolome, epigenome and immune function in order to discover new biomarkers useful for early diagnosis of autism and as potential targets for preventive personalized therapies. The European Commission has selected and financed GEMMA project as part of its Horizon 2020, the biggest EU Research and Innovation programme. The grant consists of 14,2M€ in 5-years project. GEMMA is the first project to combine a multi-omic approach with robust environmental data to exploit the analysis of the composition and function of the microbiome for personalized treatment and, ultimately, disease interception in infants at risk of is the name for a range of similar conditions, including Asperger syndrome, that affect a person’s social interaction, communication, interests and behaviour. The project will provide solid mechanistic evidence of the disease onset and progression in relation to dynamic changes in abnormal gut microbiota causing epigenetic modifications controlling gut barrier and immune functions, based on the in-depth evaluation of 600 infants at risk observed from birth and followed over time. These data will be integrated with pre-clinical studies to mechanistically link human microbiota composition/function with clinical outcome through humanized murine models transplanted with stools obtained from the ASD proband patient of recruited families. GEMMA will support novel personalized prediction (personalized treatment) and disease interception (prevention) approaches that attempt to modulate gut microbiota to re-establish/maintain immune homeostasis. The biomarkers identified in this project will contribute to a better understanding of the pathogenesis of ASD in at-risk children and the possibility to manipulate the microbiota through pre/pro/symbiotic administration for prevention and treatment, a complete paradigm shift in ASD pathogenesis and early intervention.
The identification of specific ASD metabolic phenotypes will further aid to define biomarkers that can be used as diagnostic tools and patient stratification models for other conditions in which the interplay between genome, microbiome and metabolic profile has been suspected or proved.
Finally, the project will collect biospecimens from a cohort of 600 infants as risk of ASD observed from birth, generating a unique biobank of 16,000+ blood, stool, urine and saliva samples prospectively collected that can be exploited in future multiomic studies.


State of the art

Autism Spectrum Disorders (ASD) is not a single disorder, but a spectrum of related disorders with a shared core of symptoms defined by deficits in communication, social reciprocity and repetitive, stereotypic behaviours. ASD represent a significant public health issue since it drammatically increased in the last decades reaching the prevalence of 1 in 59 children around the world (source: ADDM Network, ) with a strong sex bias (male:female=4:1). This pandemic is only partially attributable to extrinsic factors such as improved awareness and recognition and changes in clinical practice or service availability [1]. Thus, other environmental factors are likely at play to explain this pandemic. Based on the gene/environment interaction theory, several therapeutic approaches have been proposed with conflicting results [2][3][4] probably due to a multifactorial cause of ASD and the fact that ASD is a final pathological destination that can be reached through different pathways. Therefore, it is imperative to stratify the ASD population based on the identification of specific biomarkers that can assist in personalizing interventions for the most effective preventive or therapeutic results. Tackling this major public health issue requires urgent action and the formation of interdisciplinary consortia in order to further our understanding of ASD and its possible treatment/prevention. Many individuals with ASD suffer from associated co-morbidities, including seizures, sleep problems, metabolic conditions, and gastrointestinal (GI) disorders, which have significant health, developmental, social, and educational impacts. The neuroanatomical and biochemical characteristics that have been associated with autism pathogenesis [5][6][7] involve mechanisms that are direct consequences of the effects of low-grade, feverless, systemic inflammatory events [8][9] while the protective mechanisms against autism pathogenesis have strong anti- inflammatory components [10]. The gut microbiome drives immunoregulation (in particular during the 1000 days of life) and dysregulated immunoregulation [11] as well as inflammation predispose to psychiatric disorders, including autism, while psychological stress drives further inflammation via pathways that involve the gut microbiome [12][13]. Children with ASD experience four times more GI symptoms than control groups and GI symptoms may identify a unique subgroup of children with ASD [14]. Aberrant immune system activation and alterations in gut microbial composition that correlate with autism severity have also been suggested[15]. Finally, a link between gut microbiome and immune function in ASD has been reported [16] as well as improvement of both GI and behavioural symptoms in ASD subjects in which the gut dysbiosis was corrected by microbiota transfer therapy [17].


The scope of this project is to use high quality microbiome, metabolome, and other -omics data produced by large-scale international initiatives to link microbiome composition and function with specific disease for personalized prediction, prevention, and treatment of disease. Unfortunately, current large-scale datasets in ASD are mostly cross-sectional studies (cases vs controls) that do not allow the mechanistic link between microbiome composition/function with disease onset. Rather, prospective studies following infants at-risk from birth to identify potential biomarkers predictor of ASD development followed by validation on large multi-omic datasets are necessary in order to achieve this goal. The 3 patient recruitment centres will allow a global sampling based on their geographical coverage (2 European centres (Ireland and Italy) and MGH, a United States of America (US) partner) and their broad network covering over 9,500 families with ASD children, which will facilitate the recruitment of ~600 infants at risk of ASD (new-borns in families already with a child suffering from ASD). This is in line with the call’s recommendation to expand on current data in order to achieve the overall goal of this funding initiative. Therefore, the GEMMA consortia team has been designed with this approach in mind, in order to explore:
  • 1) he hypothesis that gut bacterial dysbiosis leads to epigenetic modifications, changes in metabolite profiles, increased gut permeability, increased macromolecules (food- and/or microbial-derived) trafficking and, ultimately, to altered immune responses to promote disease in a subset of individuals at-risk of ASD
  • 2) The hypothesis that the genome/metagenome interplay is responsible for the switch from immune tolerance to immune response to environmental stimuli (antigens) including dietary and microbial factors leading to neuroinflammation responsible of behavioural changes that characterize ASD and gut inflammation causing its GI co-morbidities.
If proven correct, these hypotheses support the possibility that early identification and stratification of vulnerable patients through the exploitation of gut microbiome composition/function mechanistically associated with disease pathogenesis could translate in personalized targeted interventions to prevent the onset (or attenuate the severity) of ASD and their GI comorbidities. Since microbiome changes have been associated to various human conditions including ASD this prospective cohort project (follow from birth an at-risk cohort so that the composition and function of the microbiome can be compared in the same subject before and after the onset of the disease) will test for the first time, in human clinical studies, the causality in ASD by alterations in gut microbiome composition and function mechanistically linked to the onset of the disease. The project will use a combination of patient cohorts, animal models and prospective patient samples to investigate whether specific patterns of gut dysbiosis influence epigenetic modification in the host, modifies metabolic pathways, altered gut permeability and immune response and, ultimately, trigger and/or increase the severity of ASDs and related GI comorbidities. Moreover, using specific to stratify patient populations for specific interventions, the project will attempt to validate the hypothesised mechanisms of disease via the personalized correction of gut dysbiosis and the observation of whether changes in proposed disease-specific pathophysiology take place.