Researchers at the University of California, San Diego, have used artificial intelligence (AI) algorithms to screen terabytes of gene expression data (which genes are “on” or “off” during infection) in the past. I searched for a shared pattern of pandemic patients. Viral infections such as SARS, MERS and swine flu.
The study, published June 11, 2021, revealed two distinct signatures. eBiomedicine. 1,166 pieces gene,How Human immune system Respond to Virus infectionThe second set of .20 signature genes predicts the severity of a patient’s illness. For example, you need to be hospitalized or use a ventilator.The algorithm’s utility has been validated using: Lung tissue Collected at autopsy of a deceased COVID-19 patient, infection..
“The pandemic-related signs of these viruses show how the human immune system responds to viral infections and how severe they can be. This allows for this and future pandemics. You can get a map. ” Molecular medicine At the University of California, San Diego School of Medicine and the Moors Cancer Center.
Gauche led the study in collaboration with Dr. Devasis Sahu, an assistant professor of pediatrics at the University of California, San Diego School of Medicine and an assistant professor of computer science and engineering at the Jacobs School of Engineering. Pathology, University of California, San Diego School of Medicine.
During a viral infection, the immune system releases small proteins called cytokines into the blood. These proteins Immune cells Move to the site of infection to get rid of the infection. However, sometimes the body releases too many cytokines, creating a runaway immune system that attacks its own healthy tissues. Known as a cytokine storm, this accident is thought to be one of the reasons why some patients infected with the virus (including those with normal flu) are infected and others do not die.
However, the nature, extent, and source of the deadly cytokine storm have long been unclear who is at greatest risk and how best to treat it.
“When the COVID-19 pandemic began, I wanted to leverage my computer science background to find something in common with all virus pandemics: a guide when trying to understand a new virus. I wanted to find a universal truth that could be used as a virus, “said Sahoo. “This coronavirus may be new to us, but there are limited ways our bodies can respond to the infection.”
The data used to test and train the algorithm was taken from publicly available sources of patient gene expression data (all RNA transcribed from the patient’s gene and detected in tissue or blood samples). .. Whenever a new dataset from a patient with COVID-19 became available, the team tested it with a model. They saw the same characteristic gene expression pattern each time.
“That is, what we call a prospective study, where participants participate in a study when they develop a disease and use the genetic signatures they discover to navigate unknown areas of a whole new disease. I did, “said Sahoo.
By examining the sources and functions of these genes in the first signature gene set, the study also revealed the sources of cytokine storms. White blood cells Known as macrophages and T cells. In addition, the results revealed the results of a storm. Damage to those same lung airway cells and natural killer cells, specialized immune cells that kill virus-infected cells.
“Usually, lung alveolar cells, which are designed to allow gas exchange and oxygenation of the blood, are one of the major sources of cytokine storms and therefore serve as the eye of cytokines. , We were able to show the world. Storm, “Das said. “Next, our HUMANOID Center team is modeling the human lung in the context of COVID-19 infection to investigate the effects after acute and COVID-19.”
By providing cell targets and benchmarks to measure improvement, researchers believe that this information can also help guide therapeutic approaches for patients experiencing cytokine storms.
To test their theory, the team has either a precursor version of Molnupiravir, a drug currently being tested in clinical trials for the treatment of patients with COVID-19, or a SARS-CoV-2 neutralizing antibody. Rodents were pretreated. After exposure to SARS-CoV-2, lungs cell The proportion of rodents treated with the control showed pandemic-related 166 and 20 gene expression signatures. Treated rodents do not, suggesting that treatment is effective in blunting cytokine storms.
“It doesn’t matter when the next pandemic occurs,” said Gauche, director of the Institute for Network Medicine and secretary-general of the Center for Humanoid Research at the University of California, San Diego School of Medicine. “We are building tools that are relevant not only for today’s pandemic, but for the next pandemic.”
Debashis Sahoo et al., AI-guided discovery of an invariant host response to a viral pandemic, eBiomedicine (2021). DOI: 10.1016 / j.ebiom.2021.103390
University of California, San Diego
Quote: AI predicts how patients with viral infections, including COVID-19, will be treated (11 June 2021) 11 June 2021, https://medicalxpress.com/news/ Obtained from 2021-06-ai-patients-virus-infections-covid-.html
This document is subject to copyright. No part may be reproduced without written permission, except for private research or fair trade for research purposes. The content is provided for informational purposes only.
AI predicts how patients with viral infections, including COVID-19, will progress
Source link AI predicts how patients with viral infections, including COVID-19, will progress