Periodic Reporting for period 2 - CANSAS (Clustering functional connectivity alterations in Autism Spectrum Disorders)
Reporting period: 2022-12-01 to 2023-11-30
Autism spectrum disorder is a complex neuropsychiatric condition that affects more than 1 in 100 children. This condition varies considerably across individuals. One of the most prevalent challenges in autism research is to decipher this variability and break it down into meaningful subtypes. Brain imaging methods such as rsfMRI mapping have revealed that people with autism show atypical functional communication (or “connectivity”) among brain regions. However, the origin of the brain imaging findings remains unclear. For example, the link between genetic variants associated with autism and fMRI disconnectivity is largely unknown. Using recent technology of fMRI mapping in rodents, here we identify distinct and convergent functional disconnectivity patterns across autism-relevant genetic models. These findings suggest that heterogeneous functional connectivity in ASD may encode etiologically-relevant information. By using these data, here we show that the patterns of functional dysconnectivity observed across our autism-relevant mouse models can be identified and decoded in a large collection of rsfMRI scans of individuals with idiopathic ASD.
To identify etiologically-relevant neuro-subtypes of autism, we have selected 19 mouse lines harboring a wide variety of genetic mutations associated with autism and mapped the corresponding patterns of brain activity using fMRI, a brain imaging method that can be equally applied in humans and in rodents. The use of several genetically distinct mutants allowed us to represent aspects of autism heterogeneity, while the use of rodents allowed us to map the brain imaging under the tight control of genetic and environmental factors in a way that is unattainable in human research. These controlled investigations revealed that genetic mutations associated with autism produce a range of different brain alterations. Despite this variability, we identified two dominant patterns of altered brain activity, one characterized by functional over-communication among brain regions and the other by functional under-communication. Guided by our mouse findings, we decoded rsfMRI connectivity brain maps of individuals with idiopathic autism, and we found two subtypes of participants recapitulating the dominant hypo- and hyper-connectivity brain patterns that we identified in the rodent database. We then carried out the same decoding analysis in an independent dataset of brain scans of people with autism, and we found high replicability of our subtypes. Importantly, participants in the hyper-connected subtype showed more severe socio-communicative symptoms compared to those in the hypo-connected subtype.
Very little is known about how autism can be parsed in specific subtypes. By using fMRI in rodents and humans, our work will go beyond state-of-the-art to identify biologically-meaningful neurosubtypes of autism, characterized by specific transcriptomic and behavioral signatures. The ultimate goal of our research program is to stratify and quantify the heterogeneity of autism.