Poster D82, Monday, March 26, 8:00-10:00 am, Exhibit Hall C
Dynamic transient brain networks overlap with regional gene expression in a single gene developmental disorder
Erin Hawkins1, Danyal Akarca1, Mengya Zhang1, Mark Woolrich2, Kate Baker3, Duncan Astle1; 1MRC Cognition and Brain Sciences Unit, University of Cambridge, 2Oxford Centre for Human Brain Activity, University of Oxford, 3Department of Medical Genetics, University of Cambridge
Understanding of whole-brain networks has been advanced by recent methods characterising functional connectivity. However, little is known about the dynamic nature of these networks at fast time-scales, and the underlying cellular mechanisms that drive their variability. We aimed to address this by exploring dynamic transient brain states using magnetoencephalography (MEG) in a group of individuals with the ZDHHC9 gene mutation, which affects neuronal excitability and is associated with a specific profile of cognitive difficulties. We used Hidden Markov Modelling (HMM) to explore network dynamics at rest and during an auditory oddball task in MEG, in participants with the ZDHHC9 gene mutation and age-matched controls. The HMM is a data-driven method which identifies a sequence of brain “states” corresponding to unique patterns of amplitude envelope activity which recur at different time-points, and can identify networks on a ~100ms time-scale. The HMM identified frontoparietal, frontotemporal, temporal, and visual states in the data. We then examined network dynamics using the temporal characteristics of each state: the proportion of time spent in each state, the average duration, and the number of occurrences. These characteristics distinguished the two groups on the frontotemporal and visual states. Importantly, these states also overlapped significantly with the ZDHHC9 regional gene expression profile. These results suggest that a single gene mutation can alter the dynamics of large-scale brain networks, which are tied to regionally-specific gene expression profiles. We demonstrate a valuable method for understanding this association, which may provide insights into the neural pathways linking genetic mutations to cognitive difficulties.
Topic Area: METHODS: Neuroimaging