< Back to previous page

Project

MOLECULAR DIAGNOSTICS IN AUTISM SPECTRUM DISORDER: NEXT GENERATION SEQUENCING AND FACIAL PHENOTYPING IN MEDICAL GENETICS

Despite considerable progress in the genetics of autism spectrum disorder (ASD) in the past decade, molecular diagnostics for patients with ASD remains a great challenge due to the complex genetic architecture of ASD. In clinical genetics, an etiological distinction is usually made between ASD patients with Mendelian causes (i.e. driven by a highly penetrant rare variant) and patients with multifactorial causes (i.e. driven by many genetic variants and environmental factors). Although Mendelian causes for ASD are rare, they are extremely important to diagnose in individual patients, as only they can be used for accurate genetic counseling. Therefore, the identification of highly penetrant rare variants contributing to Mendelian causes and the recognition of clinical features associated with Mendelian causes have become cornerstones in molecular diagnostics for patients with ASD. The work in this thesis is focused on improving molecular diagnostics for patients with ASD by contributing to knowledge on genotype-phenotype correlations in ASD, by bridging the gap between novel insights about the genetics of ASD and molecular diagnostics in clinical practice and by studying objective facial phenotyping as a tool to better recognize Mendelian causes of ASD.

 

With the decreasing costs of next generation sequencing (NGS), large-scale sequencing and subsequent rare variant association analyses are increasingly done to find novel risk genes for ASD and other neurodevelopmental disorders (NDDs). We participated in an international large-scale sequencing effort where targeted sequencing of 270 NDD candidate genes in over ten thousand patients with NDDs including 1894 ASD patients from Leuven was performed. Our participation contributed to rare variant association analyses, which provided evidence for the association of 148 genes to NDDs. Furthermore, the evaluation of the clinical significance of rare variants in candidate genes in individual patients from Leuven contributed to knowledge on genotype-phenotype correlations for NCKAP1, SPEN and TCF12. We also participated in a large-scale sequencing initiative where whole exome sequencing or whole genome sequencing (WGS) was performed for twenty-five thousand patients with ASD including family-based WGS for 242 ASD patients from Leuven. At the same time, NGS is also increasingly being used in the genetic workup in patients with ASD. To that end, in-house family-based WGS in 63 ASD patients from Leuven was done. Analyses of WGS data revealed several clinically significant variants in known NDD risk genes that were previously missed by chromosomal microarrays or gene panels. Furthermore, the WGS analyses revealed clinically significant variants outside known NDD risk genes, one of which allowed for the delineation of a novel NDD. This novel genetic disorder is caused by de novo missense variants in KLHL20 and is mainly characterized by mild to severe intellectual disability (ID), febrile seizures or epilepsy, ASD and hyperactivity. The genetic findings also provided useful insight in the rare variants and the clinical features that are associated with Mendelian causes of ASD. The results in this thesis showed that these rare variants mainly occur de novo and are typically found in ASD patients with ID and sporadic ASD. However, we also illustrated that these variants can be inherited from parents with mild neurodevelopmental phenotypes, frequently occur in patients with low-to-average cognitive abilities and in rare cases can even be found in families with multiple individuals with ASD.

 

A close developmental relationship between the brain and face is evident from the co-occurrence of neurodevelopmental problems, structural brain anomalies and dysmorphic facial features in patients with NDDs. During a side project in this thesis, we investigated to which extent this relationship extends to common human genetic variation by comparing genome-wide association study data of human brain and face shape. This comparison revealed 76 overlapping loci including transcription factors involved in craniofacial development, as well as members of signaling pathways implicated in brain–face cross-talk. Evidence from clinical genetics suggests that the presence of dysmorphic facial features is a strong indicator for an underlying Mendelian cause in patients with ASD. We illustrate with a case study the importance of a facial dysmorphology assessment and deep phenotyping in children with a NDD even in the absence of ID. The opinion of dysmorphologists is currently considered the gold standard to assess facial dysmorphism but this is often subject to examiner bias and is highly dependent on training and clinical experience. Therefore, we collected three-dimensional (3D) facial images of 152 patients with ASD to perform an objective analysis of facial dysmorphism. Univariate computational dysmorphism and asymmetry scores indexing unusual 3D facial development were calculated based on the 3D facial shape of ASD patients and were investigated in relation to the presence of a Mendelian cause. We found that both the computational dysmorphism scores and asymmetry scores were significantly increased in patients with Mendelian causes as compared to patient with unknown causes. Furthermore, we showed that these computational scores improved the assessment of facial dysmorphism by individual experts and thereby allowed for better recognition of Mendelian causes of ASD.

Date:1 Oct 2018 →  24 Oct 2022
Keywords:Genetics, Dysmorphology
Disciplines:Genetics, Molecular and cell biology, Medical imaging and therapy
Project type:PhD project