DNA Prediction Tools May Reveal Insights Into Fighting Genetic Diseases

Researchers from the UT Southwestern Medical Center have developed artificial intelligence (AI) algorithms capable of predicting the role and structure of DNA’s regulatory elements, which could lead to an understanding of how genetic mutations lead to disease and how scientists may be able to combat them.

According to the press release, only about 1 percent of human DNA encodes instructions for creating proteins, which play an important part in multiple DNA processes. Most other genetic material contains regulatory elements, such as promoters, enhancers, silencers, and insulators — these control how coding DNA is expressed.

However, little is known about how DNA sequence influences the function of these regulatory elements. To investigate this, the research team developed a deep-learning (DL) model known as Sei, which is designed to sort sections of noncoding DNA into 40 sequence classes. These classes categorize each snippet of DNA by “job,” such as enhancer for stem cells.

The sequence classes cover over 97 percent of the human genome. The researchers found that Sei could accurately score any sequence based on its predicted activity in each class and forecast how mutations would impact these activities.

When applied to human genomic data, the tool could characterize the regulatory architecture of 47 traits and diseases, enabling the researchers to gain insights into how mutations in these elements cause certain pathologies. This may provide researchers with a robust understanding of how genomic sequence changes are connected to disease development.

The research team then created a second AI tool, known as Orca, which can predict the 3D architecture of DNA in chromosomes based on the sequence. Following model training, Orca could accurately predict both small and large structures, including for sequences containing mutations associated with multiple health conditions, such as leukemia and limb malformation.

Using Orca, the researchers were able to generate new hypotheses about how DNA sequences may control its 3D structure. The team plans to test these hypotheses and others by using Sei and Orca to further investigate the role of genetic mutations in causing diseases, which may 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 lead researcher Jian Zhou, PhD, an assistant professor in the Lyda Hill Department of Bioinformatics at UT Southwestern, in the press release.

This research follows the publication of related research in the field of protein design.

Last month, researchers at Harvard and the University of Washington School of Medicine (UW Medicine) revealed that they had developed an AI software that can design proteins with various functions, some of which have the potential to be used to create vaccines, medicines, and other treatments. However, more testing is needed before products made using this technology become widely available.