New Delhi, August 30
A team of Indian scientists is making waves globally to develop an artificial intelligence-based system that can select carcinogenic compounds from a host of chemicals, offering huge potential to change the face of the landscape of cancer detection.
The article, “Artificial Intelligence Discovers Carcinogenic Human Metabolites,” published in the world-renowned journal “Nature Chemical Biology,” presents a new AI approach that can identify carcinogens among a sea of molecules.
The work of four collaborating institutes – Indraprastha Institute of Information Technology Delhi, IIT-Ropar, CSIR-Institute of Genomics and Integrative Biology and Rajiv Gandhi Cancer Institute – is already being hailed as a game-changer in the field of oncology just considering 5-10 percent of cancer types are hereditary and the rest are caused by exposure to carcinogens.
“Given that more than 90% of cancers are caused by exposure to compounds, the ability of our AI method to accurately identify a carcinogen can make a huge difference in cancer response. , more than 700 FDA-approved drugs had to be withdrawn because they were later found to be carcinogenic despite passing the testing stages. This is a huge burden on the industry pharmaceutical industry to determine the carcinogenic potential of a compound. Our AI system, Metabokiller, is a reliable means of accurately detecting carcinogenic compounds,” Dr. Gaurav Ahuja, lead author of the study, told The Tribune today. The Indian researchers proved the value of their AI model by flagging two compounds (which none of the currently used cancer detection models flag as carcinogens) as carcinogens and proving them to be so in experiments.
Ahuja and his collaborator, Dr. Debarka Sengupta, argue that current predictive models for predicting carcinogenicity primarily examine compounds to determine if they cause DNA damage and predict carcinogenicity based on this knowledge. “But a potentially carcinogenic compound may not damage DNA alone. Carcinogens may have many other biochemical properties that our AI model assesses for detection,” says Ahuja.
What ‘Metabokiller’ does
Metabokiller tracks biochemical properties known to be associated with carcinogens to detect cancerous compounds – a potential carcinogen could induce cell proliferation (cell division), genomic instability, anti-apoptotic response (ability of a cell not to die) , epigenetic alterations (changes in DNA) and could have an electrophilic character (where it induces a positive charge to impact DNA which is negatively charged).
Four institutes are making global waves
- Indraprastha Institute of Information Technology Delhi, IIT-Ropar, CSIR-Institute of Genomics, and Integrative Biology and Rajiv Gandhi Cancer Institute are the collaborating institutes
- The study is published in the globally recognized journal “Nature Chemical Biology” and presents a novel AI approach to identifying carcinogens among a sea of molecules.