The hereditary basis of autism investigated through matching brain networks.

The human brain is a large complex network of interconnected neurons. It is the wiring of this network that holds the key to the functioning of the brain. In this project, we aim to develop mathematical methodology to reconstruct brain networks from neuroimaging data and relate them to the brains functioning. In particular, we will develop a method to identify functional areas of the brain and match them between subjects. We will use this method to study the differences in brain networks between individuals diagnosed with autism, their unaffected siblings and controls. The novelty of our approach lies in our ability to compare individuals directly, rather than at the group level. This is essential: although typical analyses compare groups of affected subjects and controls, to analyse the hereditary basis of autism we need to compare affected subjects with their unaffected siblings at the individual level. Our aim is to find an endophenotype: a network measure that serves as an intermediate biomarker between the complex actions of genes and the heterogeneous phenotype of the disorder. Such an endophenotype would be very valuable in diagnosis and could aid in differentiating between the possibly very different neurological conditions that underlie this disorder.