Epigenetic Drug Discovery; Machine Learning’s Next Frontier
The powerful ability of Machine learning to detect the patterns in complex data is revolutionizing how we drive, how we diagnose the diseases and now it plays a role in how we discover new drugs. Researchers at Sanford Burnham Prebys Medical Discovery Institute have developed a machine-learning algorithm that could glean information from microscope images i.e, by allowing for high-throughput epigenetic drug screens that could unlock new treatments for cancer, mental illness, heart disease and more. The research study was published in eLife.
Alexey Terskikh, Ph.D., who is an associate professor in Sanford Burnham Prebys’ Development, Aging and Regeneration Program and also the senior author of the study, said in order to identify the rare few drug candidates that induce the desired epigenetic effects, scientists need methods to screen hundreds of thousands of potential compounds. He added, their study describes a powerful image-based approach that could enable high-throughput epigenetic drug discovery.
Epigenetics refers to the chemical tags on DNA that allow cellular machinery greater or less access to genes — thus altering the gene expression. Nearly all the changes in a cell, including reaction to a drug & environmental stress, are reflected by its epigenetic state. Several drugs that target epigenetic alterations are approved by the U.S. Food and Drug Administration (FDA) for the treatment of cancer, and scientists are still working to find any additional epigenetic-based therapies. Drug discovery has been slowed because of the lack of high-throughput screening methods: Researchers currently visualize the epigenetic changes using special dyes and using the traditional microscopy methods.
In this study, the researchers trained a machine-learning algorithm using a set of more than 220 drugs that are known to work epigenetically. The resulting method, Microscopic Imaging of Epigenetic Landscapes (MIEL), was successfully able to detect active drugs among them and then classified the compounds by their molecular function, spot epigenetic changes across multiple cell lines & drug concentrations, and help identify how unknown compounds work. The researchers used the approach to identify the epigenetic compounds that may be able to help treat glioblastoma, a deadly brain cancer.
Chen Farhy, Ph.D., who is a postdoctoral researcher in the Terskikh lab and also the first author of the study, said their method is ready for immediate use by pharmaceutical firms looking to develop epigenetic drug screens. Farhy added the industry and academic researchers working on mechanistic studies may also benefit from this new method, as the algorithm can detect and also categorize the epigenetic changes induced by experimental treatments, genetic manipulations or other approaches.
Terskikh and his research team are already using the machine learning algorithm to study all the epigenetic changes in aging cells, with the aim of developing compounds that can promote healthy aging — the single greatest risk factor for the diseases. This study is conducted in collaboration with Sanford Burnham Prebys professor Peter Adams, Terskikh is also eager to broaden this technology from 2D images to 3D videos, which will expand the power of the approach.