Using artificial intelligence to improve tuberculosis treatments

Medical illustration of drug-resistant Mycobacterium tuberculosis, presented in a publication of the Centers for Disease Control and Prevention (CDC) entitled, Threats of Antibiotic Resistance in the United States, 2019 (Augmented Reality Threats Report). Credit: Medical Illustrators: Alyssa Eckert; James Archer

Imagine you have 20 new compounds that have shown some effectiveness in treating a disease like tuberculosis (TB), which infects 10 million people worldwide and kills 1.5 million each year. For effective treatment, patients will need to take a combination of three or four drugs for months or even years because tuberculosis bacteria behave differently in different environments in cells — and in some cases develop to become drug-resistant. Twenty compounds in combinations of three and four drugs offer nearly 6,000 possible combinations. How do you decide which drugs to test together?

In a recent study published in the September issue of Medicine Cell ReportsIn this study, researchers from Tufts University used data from large studies containing laboratory measurements of combinations of two 12 anti-tuberculosis drugs. Using mathematical models, the team discovered a set of rules that drug pairs must meet to be good potential treatments as part of a three- and four-drug combination.

The use of drug pairs rather than measuring the three- and four-drug combination significantly reduces the amount of testing that must be performed before moving the drug combination to further study.

“Using the design rules we’ve created and tested, we can replace one drug pair with another drug pair and know with a high degree of confidence that the drug pair must work in concert with the other drug pair to kill tuberculosis bacteria in the rodent model,” says Bree Aldridge, associate professor of Molecular Biology and Microbiology at Tufts University School of Medicine and Biomedical engineering in the College of Engineering, and a faculty member in the Immunology and Molecular Microbiology program in the Graduate School of Biomedical Sciences. “The Selection process We developed it simpler and more accurate in predicting success than previous operations, which were necessarily considered fewer combinations. “

Lab Aldridge, corresponding author on the paper and associate director of the Tufts Stuart B. Levy Center for the Integrated Management of Antimicrobial Resistance, was previously developed and used Diamond, or the n-way diagonal measurement of drug interactions, a method of systematically studying pairwise and high-order drug interactions to determine the shortest and most efficient treatment regimens for tuberculosis and other potential bacterial infections. With the design rules defined in this new study, the researchers believe they can increase the speed at which scientists determine which ones medicine The most effective combinations in the treatment of tuberculosis, the second largest infectious killer in the world.

A microbiologist explains drug cocktails and how researchers find the right matches to improve results

more information:
Jonah Larkins-Ford et al, Design principles for pooling drug combinations for effective tuberculosis treatment using pairwise interpretable drug response measures, Medicine Cell Reports (2022). DOI: 10.1016 / j.xcrm.2022.100737

Introduction of
Tufts University

the quote: Using Artificial Intelligence to Improve Tuberculosis Treatments (2022, September 14) Retrieved on September 14, 2022 from

This document is subject to copyright. Notwithstanding any fair dealing for the purpose of private study or research, no part may be reproduced without written permission. The content is provided for informational purposes only.