AI and DNA predict mental health problems years after trauma


Geralt / Pixabay

Source: Geralt / Pixabay

The Biomarkers and Precision Medicine Research Center at Virginia Commonwealth University announced a new… study Published in Molecular Psychiatry that explains how to combine Artificial intelligence (AI) and genomics can produce DNA biomarkers that predict mental health problems nearly 17 years after exposure. Childhood shock.

Childhood trauma was assessed from events that meet DSM Mail-shocking Stress Criteria for the disorder in the child and adolescent Psychological assessment (CAPA) and Psychological Assessment of Young Adults (YAPA) from hundreds of children ages 9-13 who participated in the 30-year study initiated by Duke University and the North Carolina Department of Health and Human Services called the Great Smoky Mountain Study (GSMS). Blood samples and clinical data were collected in each wave.

More than 970 blood spot samples were used from more than 480 participants who provided more than 670 samples before the age of 21 years, along with a subset of more than 300 participants who submitted a sample in adulthood.

“We would expect from DNA methylation in adult outcomes,” said study lead author Edwin van den Ord, MD, a Dutch psychiatric geneticist, professor, and director at Virginia Commonwealth University’s Biomarker and Precision Medicine Research Center. We found a wide range of results as adults depressionAnd the worryAnd the alcoholicAnd the nicotine addictionpoverty, social problems and medical problems.”

Neuropsychiatric diseases and cancer have been linked to changes in DNA methylation. There are 28 million sites in the human genome where methylation can occur, according to van den Ord.

“We know where all the single nucleotide motifs are,” van den Ord said. “We take the human reference genome from the Human Genome Project and look for the CG sites, then place all the SNPs.”

Genetics The branch of biology that studies genes, genetic variation, and heredity in living organisms. DNA, deoxyribonucleic acid, is the genetic material in humans and most organisms where information is stored as a code made up of four chemical bases: adenine (A), guanine (G), cytosine (C), and thymine (T).

DNA can be modified by environmental factors, and Jenny Alteration, which can alter gene expression. DNA methylation, the process of adding methyl groups to DNA bases, is genetic modification. Given that methylation frequently occurs at CpG sites, or CG sites, the researchers identified the regions in the human genome where these sites are located. Specifically, they identified regions of DNA where the cytosine nucleotide follows the guanine nucleotide.

To identify all potential sites that could be methylated in the majority of people, the researchers began by identifying CpG sites in the human reference genome from the Human Genome Project.

“We fragment the DNA and turn it into pieces as small as 100 base pairs, and then we sequence it,” van den Ord said. “And now we know the sequence of all these little bits. Then we need to align it with the reference genome. If something goes along with a CpG-containing site, we calculate for that site how much methylation has occurred.”

Scientists calculated methylation risk scores using a synthetic intelligence (AI) Machine Learning. In AI, a linear elastic network to retreat It is a method that combines Lasso (the absolute least shrinkage and determination factor) methods and ridge regression methods.

The predictive ability of methylation risk outcomes generated by the AI ​​algorithm was “higher than that of reported trauma and cannot be explained by reported trauma, associations with demographic variables, or persistence of expected health problems from childhood to adulthood.”

According to the researchers, methylation risk scores predict a wide range of negative outcomes and have the potential to serve as a clinical biomarker for assessing health risks from exposure to trauma.

Copyright © 2022 Cami Rosso All rights reserved.


Leave a Reply

Your email address will not be published.