Medhealth Review

Finding a Way to Predict DNA Structure to Treat Disease-Causing Genetic Mutations

Artificial intelligence (AI) algorithms developed by UT Southwestern Medical Center researchers can now predict the function and structure of regulatory elements in DNA. As a result of this advancement, scientists may now have a better understanding of how genetic mutations cause disease and how to treat it.

According to the press release, only 1% of human DNA contains the instructions needed to create proteins, which are required for many DNA processes. The majority of other genetic material contains regulatory elements that control the expression of coding DNA, such as promoters, enhancers, silencers, and insulators.

However, it is unclear how the DNA sequence influences the functions of these regulatory elements. To investigate this, the researchers created Sei, a deep-learning (DL) model that can classify noncoding DNA segments into 40 different sequence categories. Each DNA fragment is classified based on its “job,” such as enhancer for stem cells.

The tool could pinpoint the regulatory architecture of 47 traits and diseases using human genomic data, providing researchers with a new understanding of the pathologies caused by mutations in these elements. For researchers, this could provide a thorough understanding of the relationship between changes in the genomic sequence and the emergence of diseases.

The research team then developed Orca, a second AI tool that predicts the 3D architecture of DNA in chromosomes based on sequence data. Orca was able to accurately predict small and large structures after model training, including sequences containing mutations linked to a variety of diseases such as leukaemia and limb malformation.

The researchers were able to generate new hypotheses about how DNA sequences might control its 3D structure using Orca. The team intends to test these and other hypotheses using Sei and Orca to further investigate the role of genetic mutations in disease development, which could lead to new treatment approaches.

“Taken together, these two programs provide a more complete picture of how changes in DNA sequence, even in noncoding regions, can have dramatic effects on its spatial organization and function,” said Jian Zhou, PhD, the lead researcher and assistant professor in UT Southwestern’s Lyda Hill Department of Bioinformatics.

Last month, researchers from Harvard and the University of Washington School of Medicine (UW Medicine) announced the creation of an artificial intelligence (AI) programme capable of designing proteins with a variety of functions, some of which could be used to develop vaccines, medications, and other treatments.

Must Read
- Advertisement -spot_img
Related Articles