Intellectual disability (or ID) is a generalized neurodevelopmental disorder characterized by different levels of intellectual impairment and adaptive malfunctioning with onset before the age of 18.

It occurs as a unique phenotype or in the context of forms of rare disease such as syndromic neurodevelopmental disorders (NDDs, e.g.,Coffin-Siris syndrome) and neurological inborn errors of metabolism (IEM, e.g., phenylketonuria). Both groups of diseases, NDDs and IEM, are good use cases for modelling phenotypic data, biomarkers, omics, imaging and pathophysiological processes associated with ID. A well-known example that connects IEM and ID is phenylketonuria, PKU, a disease that is most frequently included in new-born screening programs around the world. PKU patients are early-diagnosed and immediate introduction of the treatment enables them to have normal psychomotor development and escape intellectual disability. In total, there are 116 IEM with treatable ID (besides PKU, disorders of amino acid metabolism, congenital disorders of glycosylation, disorders of carbohydrate metabolism, etc).

In the use case we are going to combine biochemical, clinical and genetic data from early-diagnosed patients in which intellectual disability was prevented due to introduction of the early-treatment (data from newborn screening and the follow-up for patients with IEM where intellectual ability is affected) and clinical and genetic data from later-diagnosed patients in which intellectual deterioration occurred (data for patients with NDDs and IEM diseases
where intellectual ability is affected).

Creating a pool of such data from which AI will deduce the understanding of the collective characteristics of all genes associated with intellectual disability in patients with paediatric rare disease (those that were early-diagnosed
and appropriately treated and those that were late-diagnosed and late-treated/non-treated) as well as in patients with complex diseases. This innovative approach has the potential to reveal important shared biological pathways and mechanisms and help in the defining new strategies to prevent or correct intellectual disability.

Inherited Retinal Diseases (IRDs) are a group of disorders characterized by the generally progressive death or dysfunction of photoreceptors and retinal pigment epithelium (RPE) cells, leading to loss of visual function, sometimes leading to legal blindness. It is estimated that this group of diseases affects 1 in 3000 people. IRDs are clinically very heterogeneous and can be classified according to multiple parameters. In addition, IRDs present a high allelic and genetic heterogeneity.

The diagnosis of IRDs is currently based on next-generation sequencing (NGS), mainly of a panel of previously identified genes, or by studying the entire exome (whole exome sequencing; WES). Despite the undoubtedly positive impact of these technologies on the molecular analysis of diseases, the percentage of IRDs diagnoses is 50% - 70%.

An early molecular diagnosis is necessary to confirm the clinical diagnosis, offer adequate care to patients, give genetic and reproductive counselling to families, choose the most appropriate educational methods, as well as inclusion in appropriate clinical trials based on genetic information. The proposed study aims to use artificial intelligence to analyse genomic data in combination with clinical data that would allow us to reduce the genetic diagnosis time and improve the clinical management of patients, thus being able to establish personalised medicine in clinical practice.

Autism Spectrum Disorders (ASD) are neurodevelopmental disabilities characterised by social, communication and behavioural challenges. Children and adolescents with ASD are at a substantial higher risk of self-injurious and suicidal behaviour compared to the general population. The risk of suicidal behaviour and self-injury in ASD individuals is up to 9 times that of the general population. Female gender, age, cognitive rigidity hyposensitivity to pain and comorbidity have been specifically associated with suicidality and self-injury in the population group. However, the causes of the increased risk remain largely unknown and there is little knowledge about the potential role of clinical, biological, and environmental factors. There is a strong need to investigate the factors associated with suicide and self-harm in ASD individuals.

Interestingly, from a clinical point of view, ASD and suicidality (across different ICD-10 disorders) share some common aspects (reduced social abilities, rigid behaviour, ...) that may help to find the genomic and environmental underpinnings for these processes. Therefore, we will investigate phenotypic, environmental and genetic factors associated with suicidality and self-injury in a large cohort of ASD individuals with the aim to underpin the molecular processes leading to suicidal and self-harm behaviour in patients with ASD.