Researchers at the Indian Institute of Technology in Madras have come up with a new AI prediction tool that can detect cancer-causing genes in individuals.
What does this mean?
Pivot dubbed, the AI tool is based on a prediction model that uses information on mutations due to mutated gene expression, gene expression and variation in gene numbers, and confusion in biological networks.
The machine learning model is able to classify these genes as tumor suppressor genes, oncogenes or neutral genes.
A. The study, the results of which were published in the peer-reviewed journal Frontier in Genetics, found that PIVOT successfully predicted both oncogenes and tumor-suppressing genes such as TP53, and PIK3CA, and new cancer-related genes, such as PRKCA, SOX9, and PSMD4. .
Researchers have so far developed AI prognosis models for three cancers: breast invasive carcinoma, colon adenocarcinoma, and lung adenocarcinoma. According to a press release, they plan to create more models for other types of cancer.
In addition, the research team is compiling a list of cancer-causing genes to help identify appropriate drugs for patients based on their cancer profile.
Why it matters
Uncontrolled cell growth that leads to cancer is caused by oncogenes or mutations in tumor suppressor genes or both. However, according to IIT-Madras researchers, not all mutations result in cancer.
They said that identifying the genes responsible for the onset and progression of cancer in patients helps determine the appropriate combination of drugs and therapies for their recovery.
“As cancer treatment is increasingly moving towards personalized drugs, such models that make a difference in patients can be very effective,” commented Dr. Kartik Raman, research team member and associate of IIT-Madras Bhupath and Jyoti Mehta School of Biosciences. Professor 7
Although there are existing tools that can also identify cancer genes, they use unsupervised learning and predict them based on the presence and absence of mutations in that gene.
The IIT-Madras study on PIVOT, according to the research team, was the first to use supervised learning and consider the effective effects of mutations in making predictions.
Snapshots of the market
There have been recent efforts across Asia-Pacific to detect and predict cancer using AI technology. Last month, the cancer care platform Karkinos Healthcare has partnered with American oncology diagnostic firm C2i Genomics to promote AI-enabled cancer screening in India. The latter offers a diagnostic test that uses AI to identify the amount of recurrent cancer.
ICT provider Fujitsu is working on a research project with Southern Tohoku General Hospital in Japan to create an AI solution that detects pancreatic cancer from CT scans at an early stage.
A research team two years ago The National University of Singapore has come up with a cancer scorecard, a panel of 29 selected genes that help identify cancer. It was created using big data and predictive analysis of biopsies obtained from more than 30,000 patients.
“The field of precision drug research is still at an early stage. PIVOT helps push these boundaries and presents the possibility of experimental research based on identified genes,” said Malvika Sudhakar, a research team member and IIT-Madras research scholar.