New algorithm could save thousands of animals from toxic testing
After decades of animal testing for everything from industrial cleaners to beauty products, it seems as if artificial intelligence will soon be stepping into save potentially thousands of animals from further testing.
A team of researchers led by Rutgers University announced a first-of-its-kind algorithm for the testing of chemical toxicity for the benefit of workers in various industries and the animals behind the scenes. Of the 85,000 compounds used in consumer products, the majority have not been comprehensively tested for safety.
Typically, these chemicals would be tested on a range of animals. However, not only is it considered an ethical issue, but the researchers state that trying to test tens of thousands of chemicals on animals is both too costly and time consuming.
“There is an urgent, worldwide need for an accurate, cost-effective and rapid way to test the toxicity of chemicals,” said lead researcher Daniel Russo. “Animal testing alone cannot meet this need.”
Previous efforts to replace animal testing with algorithms compared untested chemicals with structurally similar compounds whose toxicity is already known. However, this creates problems as some structurally similar chemicals have very different levels of toxicity.
To overcome this, the Rutgers team’s algorithm extracts data from PubChem, a renowned database of information on millions of chemicals. The code then compares chemical fragments from tested compounds with those of untested compounds, with maths stepping in to evaluate their similarities and differences in order to predict an untested chemical’s toxicity.
To test and train the algorithm, the team took 7,385 compounds of known toxicity and presented it with 600 new compounds. For several groups of chemicals, the algorithm had a 62pc to 100pc success rate in predicting their level of oral toxicity. However, the team admitted that this algorithm does not mean it will totally replace animal testing.
“While the complete replacement of animal testing is still not feasible, this model takes an important step toward meeting the needs of industry, in which new chemicals are constantly under development, and for environmental and ecological safety,” said the corresponding author Hao Zhu.
The team’s findings have been published to Environmental Health Perspectives.
The post New algorithm could save thousands of animals from toxic testing appeared first on Silicon Republic.