Researchers at the University of Virginia have created a system that finds genetic signals of autism in brain scans with 89-95% accuracy. Called 3D transport-based morphometry (TBM), this technique uses artificial intelligence on MRI scans and models of how brain matter is distributed. It offers a more personalized way to treat autism by focusing on genetic signals rather than just behaviors. The method uses imaging to identify patterns in the brain that are linked to genetic variations associated with autism. This could help diagnose and treat autism earlier.
Autism Spectrum Disorder (ASD) is a condition that affects the way people interact with others, communicate, and behave, and is often marked by repetitive patterns of behavior.
The causes of autism are a mix of genetic factors (passed from parents to children) and environmental factors (influences of the environment over the lifespan). The diagnosis of ASD today is made mainly by observing behavior throughout the individual's development.
We know that autism has a very strong genetic basis, with up to 90% chance of being inherited. This means that it is often passed from parents to children. Therefore, a better understanding of the genes involved can help doctors identify different types of autism, discover their causes, and create more specific treatments for each person.
Despite the importance of genes, less than half of people with autism undergo genetic testing, according to researchers.
In a recent study published in the journal Science Advances, a team led by Professor Gustavo K. Rohde, from the University of Virginia, developed a system that can find genetic markers (clues) of autism in brain images. They achieved this with an accuracy of 89% to 95%.
For the study, they used magnetic resonance images of people who have genetic variations related to autism, comparing them with a control group, that is, people without these genetic variations.
To ensure that the groups were similar in other respects, control group participants were chosen based on age, gender, preferred hand use (laterality), and nonverbal IQ. In addition, people with related neurological disorders or a family history of similar problems were excluded.
The technique used in the study is called “transport-based morphometry” (TBM). It uses mathematical models to find specific patterns in the brain that are related to genetic differences called CNVs (copy number variations).
These CNVs occur when parts of DNA are duplicated or deleted, which can affect brain development. TBM helps distinguish normal variations in the brain from those associated with these genetic changes.
Scientists have recently identified more than 200 genes linked to autism, and some of these CNVs pose a significant risk. These genetic variations can arise as new mutations during the formation of eggs and sperm, or they can be inherited from parents.
A key example is a variation in the chromosomal region 16p11.2, which is linked to changes in the size and shape of the brain, as well as how gray and white matter are organized. This region is home to 29 genes and is linked to neurodevelopmental disorders such as autism.
Using the MBT technique allows scientists to see how these genetic variations affect the brain and, potentially, behavior. The study showed that different brain patterns in autism may be influenced by these variations.
The discovery paves the way for exploring new relationships between genes, brain, and behavior, not only in autism but also in other neurological disorders.
In addition, this technique could help create more personalized treatments, leading to precision medicine for autism. And this means that this truly personalized medicine could result in earlier interventions.
TBM-generated images showing spatially diffuse changes associated with 16p11.2 CNV. The 3D TBM-generated images depict physical changes in white matter (WM) and gray matter (GM) density. Red indicates a relative increase in tissue density, while blue represents a relative decrease. These findings reveal diffuse tissue overgrowth in deletion carriers and undergrowth in duplication carriers compared to controls, as highlighted by black arrows in selected regions. DOI: 10.1126/sciadv.adl5307
“We hope that the findings, the ability to identify localized changes in brain morphology linked to copy number variations, may point to brain regions and eventually mechanisms that can be leveraged for therapies,” said Rohde.
READ MORE:
Discovering the gene-brain-behavior link in autism via generative machine learning
Shinjini Kundu, Haris Sair, Elliot H. Sherr, Pratik Mukherjee and Gustavo K. Rohde
SCIENCE ADVANCES, 12 Jun 2024, Vol 10, Issue 24, DOI: 10.1126/sciadv.adl5307
Abstract:
Autism is traditionally diagnosed behaviorally but has a strong genetic basis. A genetics-first approach could transform the understanding and treatment of autism. However, isolating the gene-brain-behavior relationship from confounding sources of variability is a challenge. We demonstrate a novel technique, 3D transport-based morphometry (TBM), to extract the structural brain changes linked to genetic copy number variation (CNV) at the 16p11.2 region. We identified two distinct endophenotypes. In data from the Simons Variation in Individuals Project, the detection of these endophenotypes enabled 89 to 95% test accuracy in predicting 16p11.2 CNV from brain images alone. Then, TBM enabled direct visualization of the endophenotypes driving accurate prediction, revealing dose-dependent brain changes among deletion and duplication carriers. These endophenotypes are sensitive to articulation disorders and explain a portion of the intelligence quotient variability. Genetic stratification combined with TBM could reveal new brain endophenotypes in many neurodevelopmental disorders, accelerating precision medicine, and understanding of human neurodiversity.
Коментарі