A new algorithm could save thousands of animals from toxicity tests

Companies wishing to abandon animal testing, a new toxicity test algorithm could save thousands of lives.

After decades of animal testing for all products, from industrial cleaners to beauty products, it seems like artificial intelligence will soon save thousands of additional test animals.

A team of researchers led by Rutgers University has announced a unique algorithm to test chemical toxicity for workers in various industries and hidden animals. Of the 85,000 compounds used in consumer products, the majority have not been fully tested for safety.

As a rule, these chemicals would be tested on a range of animals. However, not only is this considered an ethical issue, but researchers say that trying to test tens of thousands of chemicals on animals is both too expensive and time consuming.

"There is an urgent need worldwide for a precise, cost-effective and timely way to test the toxicity of chemicals," said lead researcher Daniel Russo. "Animal testing alone can not meet this need."

Earlier efforts to replace animal testing with algorithms compared untested chemicals with structurally similar compounds whose toxicity is already known. However, this creates problems because some structurally similar chemicals have very different levels of toxicity.

To solve this problem, the Rutgers team algorithm extracts data from PubChem, a reputable database containing information on millions of chemicals. The code then compares the chemical moieties of the test compounds with those of the untested compounds, with the mathematicians involved evaluating their similarities and differences to predict the toxicity of an untested chemical.

To test and train the algorithm, the team took 7,385 compounds of known toxicity and presented it to 600 new compounds. For several groups of chemicals, the algorithm predicted their degree of oral toxicity between 62% and 100%. However, the team admitted that this algorithm did not mean that it would totally replace animal testing.

"Although complete replacement of animal testing is still not feasible, this model is an important step towards meeting the needs of the industry, in which new chemicals are constantly being developed, and for environmental safety. and ecological, "said author correspondent, Hao Zhu.

The results of the team were published for Environmental Health Outlook.