Scientists have developed a machine learning algorithm that helps in automating high-throughput screens of epigenetic machines.
FREMONT, CA: The powerful ability of Machine Learning (ML) has already revolutionized the way doctors drive and diagnose diseases and will continue to do so. Now, it is also transforming the way scientists are discovering new drugs. Recently, a ML algorithm has been developed by scientists at Sanford Burnham Prebys Medical Discovery Institute that gathers information from microscope images, thus enabling high-throughput epigenetic drug screens to unveil new diagnoses for heart disease, cancer, mental illness and more.
For identifying a rare handful of drug candidates that influence desired epigenetic effects, researches required new techniques to screen hundreds of potential compounds. The study describes the powerful image-based method that allows high-throughput epigenetic drug discovery.
Epigenetics are chemical tags on DNA that alter gene expression. The cell’s epigenetic state reflects all the changes in the cell, including response to a drug or environmental stress. Numerous medicines targeting epigenetic alterations have already been approved by the U.S. Food and Drug Administration (FDA) for cancer treatment. However, the growth of drug discovery has slowed down due to the lack of a high-throughput screening technique. Scientists presently visualize epigenetic changes utilizing special dyes and conventional microscopy techniques.
In the study, a ML algorithm was trained by scientists utilizing a set of beyond 220 drugs known to function epigenetically. The resulting technique, called Microscopic Imaging of Epigenetic Landscapes (MIEL), can detect active drugs, group the compounds by their molecular function, identify epigenetic changes across numerous cell lines and drug concentrations and help detect how unfamiliar compounds function. The researchers utilized the technique to identify epigenetic compounds that can help in treating glioblastoma, a fatal brain cancer.
The pharmaceutical firms searching to develop epigenetic drug screens can immediately use it. The technique will also benefit industry and academic researchers working on mechanistic studies since the algorithm can identify and categorize epigenetic changes caused by genetic manipulations, experimental treatments, or other methods.
This algorithm is already being used by Terskikh and his team to study epigenetic changes in aging cells with the objective of developing compounds that will facilitate healthy aging. The work is being conducted in amalgamation with Sanford Burnham Prebys professor Peter Adams, Ph.D. Terskikh is also very excited to widen the technology from 2D images to 3D videos for expanding the power of the approach.
Check This Out: Top Machine Learning Solution Companies