Researchers at IIT Delhi have created an artificial intelligence-based model for the detection of carcinogens in chemical compounds, which could be helpful for the pharmaceutical industry when presenting new medications. The research on the “Metabokiller” programme, according to the researchers, has also been published in Nature Chemical Biology, one of the most esteemed journals in the field of Chemical Biology.

Gaurav Ahuja, an assistant professor in the computational biology department at the Indraprastha Institute of Information Technology (IIIT) in Delhi said, “In our most recent research, an AI model was developed that could identify carcinogens (compounds that cause cancer) based on their chemical compositions. This model employs a novel strategy that focuses on the biological and chemical characteristics linked to recognised carcinogens.”

Dr. Debarka Sengupta and Mr. Ahuja have created six distinct machine learning models that thoroughly examine each query compound for characteristics linked with carcinogens. Participants in the study also came from the CSIR-Institute of Genomics and Integrative Biology in New Delhi, Rajiv Gandhi Cancer Hospital and Research Centre, and IIT Ropar.

“Notice, in the last few decades, many FDA-approved drugs have been taken back from the market since they were found to cause cancer. It also has huge importance in the cosmetics and food industries,” he added. Machine learning is a black box approach, where the cause or the reason for the prediction is largely unclear.

According to him, only 5% of cancers are inherited, whereas 95% of cancers are brought on by exposure to carcinogens (environmental substances). Mr. Ahuja stated, “We are in advanced conversation with numerous pharmaceutical companies to evaluate our software in a real-world scenario.