Researchers at the Indian Institute of Technology in Madras have devised a new AI prediction tool that can detect cancer-causing genes in individuals.
WHAT IT DOES
The AI tool, called PIVOT, is based on a prediction model that uses information about mutations, expression of genes and variation in copy numbers in genes and disruptions in the biological network due to altered gene expression.
The machine learning model can classify these genes as tumor suppressor genes, oncogenes or neutral genes.
In a studywhose findings were published in the peer-reviewed journal Frontier in Genetics, PIVOT was shown to successfully predict both oncogenes and tumor suppressor genes such as TP53 and PIK3CA, and novel cancer-related genes, such as PRKCA, SOX9 and PSMD4.
The researchers have so far developed AI prediction models for three types of cancer: invasive breast carcinoma, colon adenocarcinoma and lung adenocarcinoma. According to a press release, they plan to build more models for other cancers.
In addition, the research team is building a list of cancer-causing genes to help identify appropriate drugs for patients based on their cancer profile.
WHY IT MATTER
The uncontrollable growth of cells that leads to cancer is caused by mutations in oncogenes or tumor suppressor genes or both. However, not all mutations necessarily lead to cancer, the IIT-Madras researchers said.
They said identifying genes responsible for initiating and progressing cancer in patients helps determine the appropriate combination of drugs and therapy for their recovery.
“As cancer treatment increasingly shifts towards personalized medicine, such models that build on pinpointing differences between patients could be very helpful,” said Dr. Karthik Raman, research team member and associate professor of IIT-Madras Bhupat and Jyoti Mehta School of Biosciences.
While there are existing tools that can also identify cancer genes, these use unsupervised learning and base their prediction on the presence and absence of mutations in those genes.
The IIT-Madras study on PIVOT is the first to use supervised learning and consider the functional impact of mutations when making predictions, according to the research team.
Recent efforts have been made in Asia-Pacific to detect and predict cancer using AI technologies. Last month, platform for cancer care Karkinos Healthcare partnered with US oncology diagnosis company C2i Genomics to promote AI-based cancer monitoring in India. The latter offers a diagnostic test that uses AI to identify traces of recurrent cancer.
ICT provider Fujitsu is working on a research project with Southern Tohoku General Hospital in Japan to develop an AI solution that detects early-stage pancreatic cancer based on CT scans.
Two years ago, a research team from the National University of Singapore came up with a cancer scorecard, a panel of 29 selected genes, that helps detect cancer. It was developed using big data and predictive analysis of more than 30,000 patient-derived biopsies.
ON THE REGISTRATION
“The research field of precision medicine is still in its infancy. PIVOT helps push these boundaries and offers prospects for experimental research based on the identified genes,” said Malvika Sudhakar, research team member and a research scientist at IIT-Madras.