IIT Madras AI tool maps cancer-causing genes

Researchers at the Indian Institute of Technology, Madras (IIT Madras) have developed an artificial intelligence-based tool, PIVOT, that can predict an individual’s cancer-causing genes. Not only can this early information help prevent cancer, but it can also develop personalized cancer treatment strategies.

The prediction of cancer-causing genes is based on a model that uses information on mutations, gene expression, copy number variation in genes as well as disruptions in the biological network due to altered gene expression.

The research was led by Professor Raghunathan Ringaswamy, Dean (Global Engagement), IIT Madras, and Professor of the Department of Chemical Engineering, IIT Madras, and Dr. Karthik Raman, Associate Professor, Bhupat and Jyoti Mehta School of Biosciences, IIT Madras and a Core Member, Robert Bosch Center for Data Science and Artificial Intelligence (RBCDSAI), IIT Madras, and Malvika Sudhakar, Research Associate, IIT Madras. The results of the research were published in the peer-reviewed journal Frontier in Genetics.

Highlighting the importance of the research, Dr. Karthik Raman, Senior Member, RBCDSAI, IIT Madras said, “Cancer, as a complex disease, cannot be treated with a one-size-fits-all approach. With cancer treatment increasingly shifting towards personalized medicine, it can be Models like this that are geared towards identifying differences between patients are very useful.”

The tool is based on a machine learning model that categorizes genes as tumor suppressor genes, oncogenes or neutral genes. The tool was able to successfully predict both existing genes and tumor suppressor genes such as TP53 and PIK3CA, among others, and new cancer-related genes such as PRKCA, SOX9 and PSMD4.

Speaking about the importance of providing personalized cancer treatment, Malvika Sudhakar, a researcher at IIT Madras, said, “Research in precision medicine is still in its infancy. PIVOT helps push these boundaries and offers avenues for experimental research based on identified genes.”

It is well known that current cancer treatments are detrimental to the patient’s overall health. Knowing the genes responsible for the initiation and progression of cancer in patients can help determine the combination of drugs and treatment that is most appropriate for a patient’s recovery. Although there are tools available to identify personalized cancer genes, they use unsupervised learning and scenario prediction based on the presence and absence of mutations in cancer-related genes. However, this study is the first to use supervised learning and take into account the functional impact of mutations.

IIT Madras researchers have built artificial intelligence prediction models for three different types of cancer, including invasive breast cancer, adenocarcinoma of the colon and lung cancer. They plan to expand it to include more types of cancer. The team is also working on a list of personalized cancer-causing genes that can help determine which medications are appropriate for patients.

Leave a Comment