Depending on the unique genetic information in cancer patients' tumors, matching new treatment options, this new direction of precision medicine is often difficult to apply because it is difficult to identify all patients who may respond to a therapy. According to a recent report from the University of Pennsylvania published by Cell Reports magazine, other molecular information about patients may reveal these so-called "hidden responders." Dr. Casey Greene, assistant professor of pharmacology at the University of Pennsylvania's Perelman School of Medicine, said that targeted sequencing can identify individuals with genetic mutations that are sensitive to anticancer drugs, but many lack these mutations and improve machine learning methods. May help these patients find the right treatment. Machine learning is a branch of artificial intelligence that uses data to learn and predict. Greene and the first author of the paper, Gregory P. Way, use machine learning to classify abnormal protein activity in tumors. Their algorithm for searching for cancer genome maps (TCGA) integrates genetic data from 33 different cancer types, used by Greene and Way. The transcriptome information, that is, all messenger RNA expressed in an individual, the transcriptome information of these "hidden responders" may reveal the state of the responder. Evaluating Machine-Learning Classification of Ras Activation In particular, the researchers applied the model to the Ras pathway, which produces proteins that control cell replication and death. The normal function of the Ras protein, which causes approximately 30% of cancers, can cause cancer cells to grow and spread. Mutations are a major cause of failure of various clinical trial inhibitor drugs and vaccine treatments. This image shows machine learning and 'hidden responders'. Credit: Casey Greene, PhD Perelman School of Medicine, University of Pennsylvania, Cell Reports "This model is based on training the genetic data of human tumors in TCGA and is able to predict the response of certain inhibitors." Greene said that the role of the transcriptome in the precise treatment of tumors was underestimated, when learning with machine learning. When combined, it can help identify potential "hidden responders." Researchers are working hard to analyze uncontrolled Ras activity and tumor data through machine learning to help patients find the right treatment. ZHANGJIAGANG DINSHENGLIN TRADING CO.,LTD , https://www.dslhouse.com
Cell Reports: AI+Medical is a new trick to help identify “hidden responders†of anticancer therapies