Three DNA double helixes on a peach-colored background

Machine learning models may enable targeted gene therapy for genetic diseases

Talthough almost every cell in your body contains a copy of your gene, only a small fraction of these genes will be expressed, or turned on. This activity is controlled by special pieces of DNA called enhancers, which act as a deletion technique. This selective activation allows the cells to perform certain functions in the body, they decide to be – for example – heart cells, muscle cells, or brain cells.

However, these promoters do not always turn on the right genes at the right time, contributing to the development of genetic diseases such as cancer and diabetes. A team of biological engineers from Johns Hopkins has developed a machine learning model that can determine which enhancers play a role in normal development and disease – a development that may one day have the potential to develop therapies aimed at treating diseases by modifying genes and erasing them. by will. The results of the study appeared in the Natural genetics.

“We know that receptors control the transition between cell types for a long time, but what’s exciting about this work is that the mathematical model shows us how they can be controlled.”

Michael Beer

Professor, Whiting School of Engineering

“We know that receptors control changes between cell types for a long time, but what’s exciting about this work is that mathematical modeling shows us how they can be controlled,” said study leader Michael Beer, a professor of biological engineering. genetic medicine at Johns Hopkins University.

Human cells are very dynamic and change during our growth or in response to our environment. Beer’s team was interested in understanding how receptors affect “cell fate decisions,” or the process when one cell changes into another cell type during development. Errors in cell fate decisions are a major factor in disease development.

First, the team built a learning machine to model how genes regulate each other in a cell. From there, they used large-scale selection experiments, known as genetic screens, to identify several key genes that control cell fate decisions, as well as promoters that turn the expression of these genes on and off. Then, they used the CRISPR gene editing system to disrupt or activate potential enhancers and observe the effects on gene expression. This also allows researchers to test which factors accelerate the transformation of embryonic stem cells into endodermal cells, which is the first step in making a stomach or pancreas. Finally, the team used the data from their genetic screens to measure DNA features, such as body structure or mutation points, which are better at predicting which promoters will have the greatest impact on cell fate.

Using this new computational method, the team discovered two surprising features about the connection between the enhancer’s function and the cell’s function. First, the enhancers with the strongest effect on gene expression were all in DNA loops that enclosed the target gene—a finding that revealed precise information about how the genomic location of the enhancer helps activate the target gene. Second, the promoters only induced gene activation while the cells were changing from one type to another, and the effect disappeared as soon as the transformation was complete.

“This may explain why historically it has been very difficult to link the enhancing mutations with the associated disease. Many of these disease-related enhancers that have been identified can only change gene expression significantly when a cell transitions into a new cell type,” said Beer. .

Their results suggest that CRISPR screens designed to detect enhancers during cell state changes will have greater sensitivity. The team believes that their work can help other researchers to study the transformation of the developmental state by using models based on DNA features to predict which ones may influence the fate of the cell, and the design of experiments to identify complex enhancers by stimulating cell changes.

According to Beer, the results of the study show that the future of the cell may not be set in stone and, with further research, scientists will be able to find out which promoters are linked to certain genetic diseases, allowing them to change their function to improve prevention or treatment. genetic diseases.

“This is a new tool to study the interaction between genes and regulatory factors such as receptors, and that will provide insight into how to correct abnormal cell behavior during disease,” Beer said. “We expect our work to one day stimulate the development of treatments for cancer or other genomic diseases by targeting combinations of enhancers and CRISPR.”

Additional study authors include Jin Woo Oh, Wang Xi, Dustin Shigaki of Johns Hopkins. Other authors are from the Sloan Kettering Institute, Memorial Sloan Cancer Center, Weill Cornell Medicine, and Albert Einstein College of Medicine.

#Machine #learning #models #enable #targeted #gene #therapy #genetic #diseases

Leave a Reply

Your email address will not be published. Required fields are marked *