Genetic Analysis of Autism in Multi-Generational Families

About Genetic Analysis of Autism in Multi-Generational Families


Professor Bruce Tonge
Professor Tan
Dr Brodnicki
Dr Avril Brereton


Autism is a complex neurodevelopmental disorder characterised by significant disturbances in social, communicative and behavioural abilities.  Studies of affected twins and families with affected members have shown that genetic factors underlie disease onset.  Although certain regions in the human genome have been linked to autism, no genes have yet been identified.  This is due in part because autism is not a specific disease, but a collection of disorders, most likely with different genetic factors. Therefore, grouping together unrelated families with a broad diagnosis of autism would have limited power for gene discovery.  Unless, of course, the same gene was responsible for disease onset in the majority of families studied. Unfortunately, this does not appear to be the case as recent studies using large numbers of unrelated families have been unsuccessful for identifying autism genes.

An alternative approach is to recruit multi-generational families with a number of affected members who have strictly defined autistic characteristics.  This type of collection should provide greater statistical power for gene identification because it is more likely that a limited number of genes contribute to disease onset in these distinct families. Usually such large family pedigrees are difficult to recruit, but due to our ongoing early intervention program we have identified at least 30 families with affected members for which there is preliminary indication (eg family history) that larger family inheritance patterns are observed.  Recruitment of these families and genotyping of DNA samples from family members will provide a unique resource for identifying genes contributing to the onset of autism.


The major goal of this project is to identify and characterise autism susceptibility genes by utilising the expertise and resources available within our combined research groups.  We hypothesize that a multigenerational family based approach using strict diagnostic criteria will provide better statistical power for genetic analysis of autism compared to previous studies.  A key aspect of this approach is to recruit all family members because genetic information from unaffected individuals is as important as that obtained from those with autism.  In this regard, our specific aims are:

Stage 1

Identify those families participating in early intervention programs that have an extended family history of autism.  We will limit our recruitment to those families with multigenerational inheritance patterns for autism.

Recruit as many affected and unaffected members of identified families as possible for inclusion in this study.  (Note: Maximum participation of all family members increases the amount of genetic information resulting in greater power for genetic analysis and a higher probability of gene discovery).

Stage 2

Recruit a separate collection of unrelated, affected individuals (n>100) and their immediate families. This additional collection will be used for an association study to confirm the extended family based results and candidate gene analysis.

The clinical diagnosis of autistic disorder will be confirmed in participating individuals by qualified clinicians using a standardised assessment. Blood samples, or if not possible, cheek swabs will be collected from all participating individuals and utilised to obtain DNA samples for genotyping and genetic analysis.


DNA material has now been collected from 4 large families including members in North America and Europe.  This is very time consuming work but has established the feasibility of the project thus justifying application for research grant funding which is in progress.