New machine learning technique shows how drugs can be reused

A new machine learning method to model gene expression levels could improve the identification of genes that cause disease in humans, according to a new study from researchers at the Penn State College of Medicine. Information from the three-dimensional (3D) structure of genomes and epigenetics – how genes and environment together influence disease – allowed the researchers to identify genes associated with complex traits and diseases. These identified disease genes also help to nominate drugs that can be reused to treat new conditions.

Developing and approving new prescription drugs can be a costly and time-consuming process. However, the findings from this study may partially change that. According to researchers, instead of developing new drugs, pharmaceutical companies save time and money by reusing medications already approved by the Food and Drug Administration to treat other conditions.

The human genome is composed of genetic instructions, or DNA that is fundamental to health and disease. To carry out these instructions, DNA must be read and expressed, and gene expression is influenced by genetic variation. The same gene can be higher (or lower) expressed in people with certain mutations, which can cause disease. Scientists analyze collections of gene reads – or transcriptome – present in cells of hundreds of thousands of individuals. Transcriptome analyzes can identify genes that are differentially expressed between people with and without diseases, leading to a new understanding of the genes associated with certain conditions.

For the new data method, PUMICE (Prediction Using Models Informed by Chromatin conformations and Epigenomics), researchers at Penn State integrated transcriptomic, epigenomic and 3D genomic data using a new machine learning approach. According to the study, PUMICE was successful in identifying drugs that can reverse the expression level of disease genes and can be reused to treat various human diseases.

“Traditional approaches that analyze one drug and one disease at a time can be very inefficient,” said Dajiang Liu, co-senior author and associate professor of public health sciences and biochemistry and molecular biology at Penn State. “In contrast, a machine learning approach is based on big datalike PUMICE, could revolutionize biological and clinical research† It will significantly accelerate the process of identifying promising therapeutic targets and accelerate drug development.”

Using PUMICE, the researchers identified potential treatments for medical conditions, including COVID-19, Alzheimer’s disease, and autoimmune diseases such as Crohn’s disease, rheumatoid arthritis, ulcerative colitis and vitiligo, a skin pigmentation disorder. They noted that some of the identified drugs are already being evaluated in clinical trials, including Baracitinib, a drug for the treatment of COVID-19.

“Rediscovering drugs that are already in clinical trials demonstrates the strength of our approach,” said Bibo Jiang, co-senior author and assistant professor of public health sciences at Penn State. “We will design follow-up experiments to validate new drugs and identify the most promising for further testing in cell lines and animal models and ultimately in clinical trials†

The study is published in nature communication†

Provided by Penn State College of Medicine