Credit: Mittal, A. et al. Nat. Chem. Biol. (2022)

Metabokiller is a promising new computer-aided method to detect carcinogenic human metabolites accurately and decipher how they trigger cancers1.

Human cells are exposed to myriad metabolites, generated when the body breaks down food, drugs and other chemicals. Many of them can convert healthy cells into cancerous ones. Animal models for screening carcinogenic metabolites are expensive and time-consuming.

To find a better, faster alternative, a team at the Indraprastha Institute of Information Technology in New Delhi invented Metabokiller. It is an ensemble of six models designed to classify carcinogenic human metabolites depending on their abilities to induce oxidative stress, proliferation, genomic instability and anti-apoptosis.

The team, which also included researchers at the CSIR-Institute of Genomics and Integrative Biology in New Delhi and the Indian Institute of Technology in Punjab, used Metabokiller to virtually sift through 217, 921 metabolites stored in the Human Metabolome Database.

It predicted that about 37.5% of the metabolites were carcinogenic. Of these, 38 had been identified previously. Some of the carcinogenic metabolites were melphalan, sulphur mustard, dichloroethane, chloral and trichloroethane.

Next, the researchers validated the model’s efficiency by exposing budding yeast cells and specific human lung epithelial cells to 4-nitrocatechol and 3,4-dihydroxyphenylacetic acid, two predicted carcinogenic metabolites. Both induced malignant changes in these cells.