Adverse drug reactions (ADR) caused by medication errors are one of the most common forms of medical mistake. A new learning tool that evaluates drug interactions at the molecular level may reduce drug injury to patients.
Researchers behind a new study published in the journal EBio Medicine developed an open-source machine learning tool that could provide greater understanding of drugs associated with adverse drug reactions. In the past, our knowledge of drug toxicities and interactions has grown through observation, experimentation, and the creation and maintenance of pharma databases that capture drug toxicities.
While databases provide searchable information, the new algorithm developed in this study is able to dynamically review and compare proteins that are at the base of adverse drug reactions. In addition to providing known data, the algorithm uses artificial intelligence to compare data from multiple databases and suggest drug interactions.
For the study, researchers used a database containing information on drugs involved with ADR and a different database that identifies 184 proteins already known to interact with specific drugs. Overall, the study involved 2134 marketed drugs. Researchers created an algorithm to mesh this information and were able to identify 221 interactions between specific proteins and adverse drug reactions. These associations are useful for understanding which proteins are likely to interact poorly (and result in a bad outcome) with certain drugs.
Rather than stand as a new static resource for Pharma development, the new tool can be shown new information on which to build and compare reactions and interactions between compounds already evaluated by the tool, and suggest how they might interact with human biology for the better or the worse.
Notes study co-author, Robert Ietswaart, “Machine learning is not a silver bullet for drug discovery, but I do believe it can accelerate many different aspects in the difficult and long process of developing new medicines.”
In addition to identifying potential candidates for adverse reactions, the tool could make experimental drugs safer in human clinical trials. Ietswaart explained, “This could reduce the risks that study participants face during the first in-human clinical trials and minimize risks for patients if a drug gains FDA approval and enters clinical use.”
Machine-learning may provide a new tool for an old problem. Safer pharmacology means safer medical care—and possibly less danger each time you pick up a prescription at the pharmacy.
Trusted Washington, D.C. and Maryland attorneys deliver aggressive legal representation on your behalf
The attorneys at Schochor, Staton, Goldberg, and Cardea, P.A. deliver strong legal support to those who suffer medical malpractice or negligent injury in Maryland, Washington, D.C., or across the nation. We have obtained more than $1 billion in compensation for our injured clients in the past 36 years. Contact us or call 410-234-1000 to schedule a free consultation to discuss your injury.