In a new development, an algorithm developed by researchers helps to predict the side effects of drugs before they are commercialized. The algorithm developed by a team of researchers at Royal Holloway, University of London works on the same principle using which movies are recommended to users.
The machine learning approach developed by two researchers at the Department of Computer Science, Royal Holloway is the first of its kind in the field. It can predict the percentage of population that will be affected by a specific side effect of a drug.
While, at present, during clinical trials, numerous side effects of medical drugs are not observed. The side effects are identified only after the drug has reached patients, which can lead to significant morbidity and mortality in populations.
Algorithm similar to one to Recommend Movie for Users
The algorithm is similar as the one used for Netflix. To recommend movies, the algorithm predicts users taste and then suggests movies to match their taste. In the same way, the new algorithm will help to determine the reaction induced of a drug, and the percentage of people who will have side effects after the first stages of human trials. Using this data, this will direct clinical trials for the drug going forward.
“The frequency of side effects of drugs after first stages of clinical trials is extremely important to be predicted,” said one of the researchers behind the development. Presently, there is no such system that can do this.
Meanwhile, such a development is vital for patient care as well as the pharmaceutical industry. For clinical practice, the accurate estimation of frequency of side effects of drugs is important for patient care. And, at the same time, it is essential for pharmaceutical companies to reduce the risk of drug recall from the market, or expensive reassessment of frequency of side effect through new clinical trials.