Nat Commun: Machine learning aids drug reuse to treat Alzheimer's disease

As the global population ages, Alzheimer's disease (AD) has become a growing medical crisis, with the main risk factor being increased life expectancy.
estimates that the incidence of the disease will more than double in the coming decades in the absence of effective prevention and treatment programmes.
in addition to the direct impact on human health and well-being, long-term care for affected individuals also poses a significant financial burden.
are working to develop AD's disease therapy, including about 200 clinical trials to date, but the results have been largely negative, with many failures due to lack of efficacy or toxicity.
failure of clinical trials of each new molecular entity (NME) consumes a lot of time and resources.
, the re-use of fda-approved drugs for different adaptations is less costly, more specific about possible toxicity, and has a higher success rate (30 percent) than developing NME.
recently, researchers published a paper in the journal Nature Medicine, presenting DriaD, a machine learning framework that quantifies the potential link between pathology and molecular mechanisms of the AD severity (Braak phase).
80 FDA-approved and clinically tested drugs were used in differentiated human nerve cell cultures and collected a list of genes produced by their disturbances, driad was used to analyze the list of genes to produce a list of possible re-use candidates.
then examined the highest-scoring drugs for common trends among their targets.
DRIAD method may be used to nominate drugs that can be easily evaluated in clinical trials after additional validation and identification of associated drug markers.