PLOS Computational Biology: AI Fusion Genomics research predicts early molecular markers of Alzheimer's disease.

Introduction: Next-generation sequencing (NGS) technology has become a powerful tool for dissecting the molecular and pathological characteristics of various human diseases, however, the finiteness of biological samples from different stages of disease is a major obstacle to the study of disease progression and the identification of early pathological changes.
recently, researchers at the Korea Institute of Brain Sciences used a deep learning technique to predict early molecular markers of Alzheimer's disease (AD) using artificial intelligence (AI) fusion genomics research.
years, deep learning technology has been driving the artificial intelligence (AI) technology revolution.
the next generation of sequencing (NGS) technological advances, histologically based research data has become easily accessible and standardized.
, a study by the Korea Institute of Brain Science analyzed cerebral cortological tissue data from the Alzheimer's disease (AD) mouse model using advanced adversive generation networks (GANs) deep learning techniques.
found that beta amyloid protein increased in the early stages of the disease and altered the biosynthesis of cholesterol.
the study was published in PLOS Computational Biology, an international academic journal in the field of computational biology.
is entitled "A practical application of generative adversarial networks for RNA-seq analysis to predict the molecular progression of Alzheimer's disease" beta amyloid protein is a protein that causes Alzheimer's disease (AD).
in the normal brain, if excessive accumulation, will be removed by small glial cells and so on.
cholesterol must also be maintained at a certain level in the blood to form the cell membrane, regulate membrane mobility and maintain balance in the body.
if the above process does not work properly, pathological abnormalities will occur in the body.
team analyzed data on the cortic tissue of the AD mouse model using advanced adversive generation networks (GANs) for deep learning techniques.
GANs is an algorithm that generates data through competition between generators and identifiers, analyzes the generated data, and generates composite data close to real images.
, GANs technology has been used to make videos of President Barack Obama's fake speeches and can be used to predict facial aging.
overview of the application of GAN in a large number of RNA-seq data.
: The team at the Korea Institute of Brain Sciences used GANS to simulate AD gene expression in mice and observe the process of gene expression changing from normal to AD state.
results found that beta amyloid increased levels in the early stages of the disease and altered the biosynthesis of cholesterol.
findings have also been confirmed by RNA sequence analysis of brain tissue after death.
this means that the increase in beta amyloid proteins triggers biosynthesis of cholesterol, and that the combination of these two processes is likely to affect synoptic formation and synoptic plasticity through interactions.
This study is based on a unique research technique that integrates bioininsergy with AI, and its significance is to provide researchers with more systematic analysis and experimental design, which represents a new way to predict biological changes in the early stages of disease development and can be applied to the healthcare industry.
transition curves and heat maps of genes associated with cholesterol biosynthesis and cholesterol metabolism.
: "GANs are a useful tool for analyzing differences in gene expression caused by disease and for analyzing molecular progression by explaining the causes of the phenomenon," said Dr. Cheon of the Korea Institute of Brain Sciences.
this approach and the accumulation of histological data are expected to help us overcome the greatest limitations of analysis related to brain disease and aging, a time-consuming process of obtaining samples.
research is an exclusively funded project by KBRI.
.