New AI tools accurately predict role of DNA regulatory elements

According to two recent studies conducted in Natural genetics. These tools could eventually shed new light on how genetic mutations lead to disease and could lead to a new understanding of how genetic sequence influences the spatial organization and function of chromosomal DNA in the nucleus, said study author Jian Zhou, Ph.D., assistant professor. in Lyda Hill’s Bioinformatics Department at UTSW.

Taken together, these two programs provide a more complete picture of how changes in DNA sequence, even in non-coding regions, can have dramatic effects on its spatial organization and function.”

Jian Zhou, Assistant Professor, Lyda Hill Department of Bioinformatics, UT Southwestern Medical Center

Dr. Zhou is also a Fellow of the Harold C. Simmons Comprehensive Cancer Center, a Lupe Murchison Foundation Fellow in Medical Research, and a Fellow of the Cancer Prevention and Research Institute of Texas (CPRIT).

Only about 1% of human DNA encodes instructions for making proteins. Research over the past few decades has shown that much of the remaining non-coding genetic material contains regulatory elements – such as promoters, enhancers, silencers and insulators – that control how coding DNA is expressed. . How the sequence controls the functions of most of these regulatory elements is not well understood, Dr. Zhou explained.

To better understand these regulatory components, he and his colleagues at Princeton University and the Flatiron Institute developed a deep learning model they named Sei, which precisely sorts these non-coding DNA snippets into 40 “sequence classes” or uses – for example, as an enhancer for gene activity in stem cells or brain cells. These 40 sequence classes, developed from nearly 22,000 datasets from previous studies investigating genome regulation, cover more than 97% of the human genome. Additionally, Sei can score any sequence by its predicted activity in each of 40 sequence classes and predict the impact of mutations on those activities.

By applying Sei to human genetic data, the researchers were able to characterize the regulatory architecture of 47 traits and diseases recorded in the UK Biobank database and explain how mutations in regulatory elements cause specific pathologies. Such abilities can help gain a more systematic understanding of how genomic sequence changes relate to diseases and other traits. The results were published this month.

In May, Dr. Zhou reported on the development of a different tool, called Orca, which predicts the 3D architecture of DNA in chromosomes based on its sequence. Using existing datasets of DNA sequences and structural data derived from previous studies that revealed the molecule’s bends, twists and turns, Dr. Zhou trained the model to make connections and assessed the model’s ability to predict structure at different length scales.

The results showed that Orca predicted small and large DNA structures based on their sequences with high accuracy, including for sequences carrying mutations associated with various health conditions, including a form of leukemia and malformations of the members. Orca also allowed researchers to generate new hypotheses about how DNA sequence controls its local and large-scale 3D structure.

Dr. Zhou said he and his colleagues plan to use Sei and Orca, both of which are publicly available on web servers and as open source code, to further explore the role of genetic mutations in molecular manifestations and physical diseases – research that could eventually lead to new ways to treat these conditions.

The Orca study was supported by grants from CPRIT (RR190071), the National Institutes of Health (DP2GM146336), and the UT Southwestern Endowed Scholars Program in Medical Science.


UT Southwestern Medical Center

Irene B. Bowles