Monday, October 18, 2021

    Latest Posts

    Scientists develop artificial intelligence method to predict anti-cancer immunity

    Activation of the T cell immune response by the interaction of MHC-II (red) with the T cell receptor (TCR, blue) and CD4 (light blue). CD4 and MHC-II are proteins expressed by T cells and antigen-presenting cells, respectively, and support TCR recognition of antigens.Credit: UT Southwestern Medical Center

    Researchers and data scientists at the UT Southwestern Medical Center and MD Anderson Cancer Center have developed artificial intelligence technology that can identify the recognition of cell surface peptides produced by cancer cells, called new antigens, by the immune system. bottom.

    pMTnet technology, details online Nature Machine Intelligence, May lead to new ways of predicting cancer prognosis and potential responsiveness to immunotherapy.

    “It seemed like an impossible feat to determine which new antigens bind to T cell receptors and which ones don’t, but machine learning is making progress.” , Senior author Dr. Tao Wang, assistant professor of population, said. And Data Science, and Harold C. Simmons Center for Comprehensive Cancer and UT Southwestern Host Defense Genetics Center.

    Genome mutation cancer cell Display different new antigens on their surface.Some of these new antigens are recognized by immune T cells looking for signs of cancer or foreign invaders, and cancer cells Immune system.. However, others appear invisible to T cells, allowing the cancer to grow unsuppressed.

    “For the immune system, the presence of new antigens is one of the biggest differences between normal and normal. Tumor cellsTianshi Lu, the first co-author with ZeZhang, a doctoral student at Tao Wang Labs, said that tumor immunology is a therapeutic response to tumorigenesis, metastasis, prognosis, and various cancers. If we can understand what stimulates it, we may be able to use this knowledge in a variety of ways to fight cancer, “Lou said.

    Predicting which new antigens will be recognized by T cells allows researchers to develop personalized cancer vaccines, design better T-cell-based therapies, and allow patients to develop other types of other types of treatments. It helps to predict how well it responds to immunotherapy. However, there are tens of thousands of different new antigens, and methods of predicting which one triggers a T cell response have proven to be time consuming, technically difficult, and costly.

    With grants from the National Institutes of Health (NIH) and the Texas Cancer Prevention Institute (CPRIT), the research team is looking for better technology. Machine learning..They trained Deep learningA base algorithm named pMTnet using data from known combined or unbound combinations of three different components. A protein called Major Histocompatibility Complex (MHC) cancer Cell surface; and T cell receptor (TCR) involved in recognition Neoantigen-MHC complex. The algorithm was then tested against datasets developed from 30 different studies that experimentally identified bound or unbound new antigen T cell receptor pairs. This experiment showed that the new algorithm has a high level of accuracy.

    Researchers can use this new tool Cancer Genome Atlas, A public database that holds information from over 11,000 primary tumors. pMTnet has shown that new antigens generally elicit a stronger immune response compared to tumor-related antigens. We also predicted which patients responded better to immune checkpoint inhibition therapy and showed better overall survival.

    “As an immunologist, the most important hurdle facing immunotherapy today is the ability to determine which antigen is recognized by which T. cell Corresponding author Dr. Alexander Ruben, an assistant professor of thoracic, head and neck oncology at MD Anderson, said in order to utilize these combinations for therapeutic purposes, “pMTnet surpasses current options and is significantly closer to this purpose. . ”

    Scientists identify new targets for cancer vaccines

    For more information:
    Tianshi Lu et al, T cell receptor-deep learning-based prediction of antigen binding specificity, Nature Machine Intelligence (2021). DOI: 10.1038 / s42256-021-00383-2

    Quote: Scientists obtained anti-cancer immunity from on September 23, 2021 (2021). We are developing an artificial intelligence method for predicting (September 23)

    This document is subject to copyright. No part may be reproduced without written permission, except for fair transactions for personal investigation or research purposes. The content is provided for informational purposes only.

    Scientists develop artificial intelligence method to predict anti-cancer immunity Source link Scientists develop artificial intelligence method to predict anti-cancer immunity

    The post Scientists develop artificial intelligence method to predict anti-cancer immunity appeared first on California News Times.

    Latest Posts

    Don't Miss

    Stay in touch

    To be updated with all the latest news, offers and special announcements.