Study: Artificial Intelligence Unveils the Gene of "Nervous Growth Disorders"

Technical researchers have developed artificial intelligence that contribute to the acceleration of genes that play a role in nervous growth disorders, such as autism spectrum disorder, epilepsy and delayed growth. This new computer instrument helps to draw an extensive genetic map of these disorders, contributing to achieving an accurate diagnosis and understanding of the disease manisms, in addition to developing treatments that target genetically. Despite the great progress of the discovery of genes associated with nervous growth disorders, many patients do not receive a clear genetic diagnosis, which is an indication of the presence of other genes that have not yet been identified. An innovative approach published by the researchers in the study in the American Journal of Human Genetics (AJHG) has used an innovative approach on the basis of the benefit of artificial intelligence to analyze genetic patterns related to nerve growths, rather than limiting the direct comparison of individuals with these disorders. Scientists have used the study of genes previously identified as related to these cases, and then used artificial intelligence models to discover common dividers, such as genetic expression patterns and the degree of bearing genetic mutations and reactions in different biological paths. Through this methodology, the team could develop models that are able to predict other genes that can play a role in the diseases themselves, even if it has not previously been identified by traditional methods. These models have proven great accuracy in the classification of genes that are likely to be responsible for the autism spectrum disorder, delayed growth, as it has shown that the predictive genes have a high probability for involvement in these disorders compared to other genes, which open the way for new discoveries that can improve the accuracy of the diagnosis. The lead author of the study, “Ryan Dindsa”, a professor of pathology and immunity to the Baylor College of Medicine, said that the team studied genetic expression patterns at the level of one cell in the human brain during the growth stages. The results showed that the models trained in this data can predict high accuracy with genes associated with autism spectrum disorder and delayed growth and epilepsy. But the team was not satisfied with it, but rather tried to improve these models by integrating more than 300 extra biological features. Smart models with highly predictive predictive to achieve greater accuracy in predicting genes associated with nervous growth disorders, researchers have improved artificial intelligence models by integrating more than 300 additional biological properties, which distinguishes the ability of models to distinguish between genes related to these disorders, and what is not related to them. These properties have included the extent of genes tolerance of mutations, that is, their sensitivity to genetic changes that can lead to disease, in addition to analyzing their interactive relationship with other genes known to cause neurological disorders, as well as their functional roles in different biological paths. According to the study, models became more accurate in classifying genes at high risk, as the possibility of detecting genes associated with neurological disorders significantly increased, making this technique a powerful tool to support future genetic studies and improve diagnostic and treatment opportunities. Dinda emphasized that these models have a high prediction capacity because it shows that the genes classified in higher ranks, in some cases, the traditional genetic standards weigh up to 6 times. He explained that some of the genes identified by these models were more likely to agree with previous studies at a rate of between 45 and 500 times compared to the less -ranged genes. The researchers believe that these models will work as analytical instruments that help verify the genes found in genetic studies, but that they have not yet obtained sufficient statistical evidence to prove their relationship to nerve growth disorders. The team hopes that this technique will contribute to the acceleration of the process of detecting genes and achieving the early diagnosis of patients, with its effectiveness to judge in the future in detail.