[ad_1]
The use of an animal to test the toxicity of chemicals may become an outdated day thanks to a low-cost, high-speed algorithm developed by researchers at Rutgers and other universities.
Toxicity tests – determining the degree of exposure to a chemical hazardous to humans – are essential for the safety of millions of workers in various sectors. But of the 85,000 compounds used in consumer products, the majority have not been fully tested for safety. Animal testing, in addition to their ethical concerns, may be too expensive and take too long to meet this need, according to the study published in Environmental Health Outlook.
"There is an urgent need, all over the world, for a precise, economical and quick way to test the toxicity of chemicals, to ensure the safety of the people who work with them and the environment in which they are used, "said Daniel Russo, senior researcher, PhD student at the Center for Computational and Integrative Biology Rutgers University-Camden. "Animal testing alone can not meet this need."
Previous efforts to solve this problem used computers to compare untested chemicals with structurally similar compounds whose toxicity is already known. But these methods were unable to evaluate structurally unique chemicals – and were confounded by the fact that some structurally similar chemicals had very different levels of toxicity.
The group led by Rutgers has addressed these challenges by developing a unique algorithm that automatically extracts data from PubChem, an information database of the National Institutes of Health on millions of chemicals. The algorithm compares the chemical fragments of the tested compounds with those of the untested compounds and uses several mathematical methods to evaluate their similarities and differences in order to predict the toxicity of an untested chemical.
"The algorithm developed by Daniel and the Zhu lab exploits a huge amount of data and helps discern the relationships between fragments of compounds of different chemical clbades, exponentially faster than that of a human," said co-author Lauren Aleksunes, badociate professor at Rutgers's Ernest Mario School of Pharmacy, and the Rutgers Institute of Occupational Health and Environmental Sciences. "This model is effective and provides companies and regulators with a tool to prioritize chemicals that may require more comprehensive testing of animals prior to commercial use."
To refine the algorithm, the researchers started with 7,385 compounds for which toxicity data are known, and compared them with data on the same chemicals in PubChem. They then tested the algorithm with 600 new compounds. For several groups of chemicals, the algorithm led by Rutgers predicted their level of oral toxicity between 62% and 100%. And by comparing the relationships between sets of chemicals, they have brought to light new factors that can determine the toxicity of a chemical.
Although the algorithm only aimed to badess the level of toxicity of chemicals when they were consumed orally, Rutgers-led researchers conclude that their strategy can be extended to predict other types of toxicity.
"While the complete replacement of animal testing is still not feasible, this model represents an important step toward meeting the needs of the industry, in which new chemicals are constantly developing, as well as for environmental and ecological security, "said author correspondent Hao Zhu. Associate Professor of Chemistry, Rutgers-Camden and Rutgers, and the Institute of Environmental and Occupational Health Sciences.
###
Wenyi Wang, recent graduate of the Computational and Integrative Biology Center at Rutgers-Camden; Sunil Shende, graduate director of the Rutgers-Camden Computer Department, affiliated with the Center for Computational and Integrative Biology; and researchers from the Integrated Laboratory Systems, the Johns Hopkins Bloomberg School of Health and the University of Kostanz.
Warning: AAAS and EurekAlert! are not responsible for the accuracy of the news releases published on EurekAlert! contributing institutions or for the use of any information via the EurekAlert system.
Source link
