Advances in network neuroscience: towards connectome-based predictions and mechanisms

Neurodegenerative diseases involve accumulation of aberrant proteins in the brain, leading to brain damage and progressive cognitive and behavioral dysfunction. These diseases, including Alzheimer’s disease, Parkinson’s disease, and frontotemporal dementia spectrum disorders, represent the primary cause of dementia in older people. Many gaps exist in our understanding of how these diseases initiate and how they progress through the brain. However, evidence has accumulated supporting the hypothesis that aberrant proteins can be transported using the brain’s intrinsic network architecture — in other words, using the brain’s natural communication pathways. This theory forms the basis of connectome-based computational models, which combine real human data and theoretical disease mechanisms to simulate the progression of neurodegenerative diseases through the brain.
Connectome-based disease progression allow testing of mechanistic hypotheses of disease biology, but in the context of humans, where experimentation is often not possible. However, the potential of these models extends beyond hypothesis testing. In combination with emerging longitudinal data and advances in mesoscale molecular brain mapping, connectome-based models are poised to achieve clinically useful individual-level predictions, as well as to generate novel biological insights into disease progression. In this talk, I will highlight recent work by my lab and others that is already moving the needle toward these lofty goals.
First, I will discuss work using connectome-based models to test hypothetical modes of disease progression across several neurodegenerative diseases, including prediction of tau progression in Alzheimer’s disease. Second, I will touch on work using connectome-based models to forecast future progression of neuropathological and cognitive outcomes, highlighting their potential utility in current and upcoming clinical trials. Third, I will showcase novel efforts to model multiple disease elements simultaneously — in this case, accumulation of beta-amyloid, tau and neurodegeneration in Alzheimer’s disease. Finally, I will end by describing work combining connectome-based models with novel human brain mapping data to gain new insights into neurodegenerative disease biology. Altogether, this talk will summarize the state-of-the-art in disease progression modeling using connectome-based models, with a focus on novel efforts toward clinical utility and novel hypothesis generation. 

In complex networks, like the brain, information transfer is ensured by an efficient topological architecture. A methodology to causally validate this structure-function relationship is represented by Transcranial Magnetic Stimulation (TMS), a form of noninvasive brain stimulation that can act on the neural mechanisms underpinning cognition. However, the vast majority of approved interventions and approaches still rely on anatomical landmarks and rarely on the individual structure of networks in the brain, drastically reducing the potential efficacy of neuromodulation. In trying to fill this gap, we employed an approach to study the topographical properties of the individual connectome, determine its response to external perturbations and use this information to tailor the selection of stimulation targets in the brain by relying on a target search algorithm leveraging on mathematical tools from Network Control Theory (NCT) and whole brain connectomics analysis. By means of computational simulations, we identified the optimal stimulation target(s)d at the individual brain level capable of reaching maximal engagement of the stimulated networks’ nodes. At the model level, in silico predictions suggested that stimulation of NCT-derived cerebral sites might induce significantly higher network engagement, compared to traditionally employed neuromodulation sites, demonstrating NCT to be a useful tool in guiding brain stimulation. In conclusion, the use of NCT to computationally predict TMS pulse propagation suggests that individualized targeting is crucial for more successful network engagement. Future studies will be needed to verify such prediction in real stimulation scenarios. 

Caio Seguin will demystify some misconceptions related to the use of fMRI connectivity to infer underlying patterns of brain activity and brain communication. Specifically, empirical evidence of topographic convergence between structural and functional connectivity has prompted the widely held assumption that local axonal output drives fMRI connectivity. Within this framework, reduced or increased activity in a brain region should thus result in reduced (hypo-) or increased (hyper) connectivity with the region's targets. This talk will illustrate the result of recent chemogenetic and electrophysiological studies in mice that challenge this simple framework. Specifically, he will show how fMRI hyper- and hypoconnectivity may in fact counterintuitively reflect reduced and increased cortical activity, respectively. This updated framework may offer novel opportunities to biologically decode fMRI (dys)connectivity in human disorders. 

The dominance of fMRI as the primary lens through which the functional connectome is studied has led to the implicit assumption that the connectome constitutes a single stream of time-varying connectivity patterns. However, a multi-modal approach challenges this view. Specifically, source-space EEG-derived connectivity in a large cohort demonstrates that sub-second connectome dynamics are methodologically reliable, individually specific, highly heritable, and predictive of cognitive abilities. Sadaghiani will argue that these rapid dynamics are neurally largely distinct from the dynamics observed in fMRI. This conclusion is derived from concurrently recorded source-localized scalp EEG-fMRI and human intracranial EEG-fMRI. Specifically, spatially similar connectome states (recurrent connectivity patterns) are found in fMRI and all EEG frequency bands, but these states occur at different timepoints in fMRI and each EEG band. This multiplex of temporally distinct but parallel connectome trajectories can enable concurrent communication across multiple sets of brain regions in an independent manner.