The Indian Institute of Technology (IIT) Madras researchers have formulated an artificial intelligence-based mathematical model. The AI algorithm will spot cancer-causing changes in cells. This algorithm studies DNA structures to characterize the genetic changes responsible for cancer progression, which are responsible for causing cancer.
Not much work has been done on this technology so far. Understanding the underlying mechanisms of these changes will help the specialist clinicians to choose the exact cancer treatment strategy for the patient. This claim has been made in an article published in the International Cancer Journal.
One of the major challenges experienced by cancer researchers is driver mutations, due to the relatively small number and several such passenger mutations, Professor B Ravindran, Head of IIT Madras’s Robert Bosch Center for Data Science and AI (RBCDSAI), said in a statement. The difference between the large numbers is included, which does not affect whether the cancer is progressing.
In this study, they have decided to look at this problem from a different viewpoint. The main goal was to discover patterns in DNA sequences. The underlying hypothesis was that these patterns would be unique to the different types of mutation – drive, and passenger, and could therefore be mathematically modeled to differentiate between the two classes.
Table of Contents
Using simple AI techniques, the researchers developed a new prediction algorithm, NBDriver, and tested its performance on several open-source cancer mutation datasets. Professor B Ravindran explained that our model can differentiate between well-studied drivers and passenger mutations from cancer genes with an accuracy of 89 percent.
Furthermore, combining predictions from NBDriver and three other commonly used driver prediction algorithms yielded an accuracy of 95 percent, significantly outperforming existing models.
Furthermore, Karthik Raman, Associate Professor, Department of Biotechnology, IIT Madras, told NBDriver that 85 percent of patients diagnosed with glioblastoma multiforme (GBM), a particularly aggressive type of cancer affecting the brain or spine, are rare. The driver can accurately identify the mutation.