Summary: A new deep learning algorithm that looks for the burden of genomic variants is 70% accurate in identifying specific mental health disorders within the African-American community.
Fountain: CUT
Historically, minority populations have been underrepresented in existing studies addressing how genetic variations may contribute to a variety of disorders. A new study from researchers at Children’s Hospital of Philadelphia (CHOP) shows that a deep learning model has promising accuracy when helping to diagnose a variety of common mental health disorders in African-American patients.
This tool could help distinguish between disorders, as well as identify multiple disorders, encouraging early intervention with greater precision and allowing patients to receive a more personalized approach to their condition.
The study was recently published in the journal Molecular Psychiatry.
Proper diagnosis of mental disorders can be challenging, especially for young children who are unable to complete questionnaires or rating scales. This challenge has been particularly acute in understudied minority populations. Previous genomic research has found several genomic signals for a variety of mental disorders, with some serving as potential therapeutic drug targets.
Deep learning algorithms have also been used to successfully diagnose complex diseases such as attention deficit hyperactivity disorder (ADHD). However, these tools have rarely been applied in large African-American patient populations.
In a single study, researchers generated whole genome sequencing data from 4,179 blood samples from African-American patients, including 1,384 patients who had been diagnosed with at least one mental disorder. This study focused on eight common mental disorders, including ADHD, depression, and anxiety. , autism spectrum disorder, intellectual disability, speech/language disorder, developmental delays, and oppositional defiant disorder (ODD).
The long-term goal of this work is to learn more about the specific risks of developing certain diseases in African-American populations and how to potentially improve health outcomes by focusing on more personalized treatment approaches.
“Most studies focus on only one disease, and minority populations have been grossly underrepresented in existing studies using machine learning to study mental disorders,” said senior author Hakon Hakonarson, MD, Ph.D. ., Director of the Applied Research Center. Genomics in CHOP.
“We wanted to test this deep learning model in an African-American population to see if it could accurately differentiate patients with mental disorders from healthy controls, and if we could correctly label disorder types, especially in patients with multiple disorders.”
The deep learning algorithm searched for the loading of genomic variants in coding and non-coding regions of the genome. The model demonstrated an accuracy of over 70% in distinguishing patients with mental disorders from the control group. The deep learning algorithm was equally effective in diagnosing patients with multiple disorders, with the model providing exact diagnostic matches in approximately 10% of cases.
The model also successfully identified multiple genomic regions that were highly enriched for mental disorders, meaning they were more likely to be involved in the development of these medical disorders. Biological pathways implicated included those associated with immune responses, antigen and nucleic acid binding, a chemokine signaling pathway, and guanine nucleotide-binding protein receptors.
However, the researchers also found that variants in non-protein-coding regions seemed to be implicated in these disorders more frequently, meaning they may serve as surrogate markers.
“By identifying genetic variants and associated pathways, future research aimed at characterizing their function may provide a mechanistic view of how these disorders develop,” said Hakonarson.
About this research news on deep learning, genetics and mental health
Author: press office
Fountain: CUT
Contact: Press Office – CHOP
Image: The image is in the public domain.
original research: Open access.
“Application of a deep learning algorithm on whole genome sequencing data uncovers structural variants associated with multiple mental disorders in African-American patientsby Yichuan Liu et al. Molecular Psychiatry
Abstract
See also
Application of a deep learning algorithm on whole genome sequencing data uncovers structural variants associated with multiple mental disorders in African-American patients
Mental disorders present a global health problem, while diagnosing mental disorders can be challenging. Diagnosis is even more difficult for patients who have more than one type of mental disorder, especially for young children who are unable to complete standardized diagnostic questionnaires or rating scales. In the last decade, multiple genomic association signals have been reported for mental disorders, some of which present attractive drug targets.
At the same time, machine learning algorithms, especially deep learning algorithms, have been successful in diagnosing and/or labeling complex diseases, such as attention deficit hyperactivity disorder (ADHD) or cancer. In this study, we focus on eight common mental disorders, including ADHD, depression, anxiety, autism, intellectual disabilities, speech/language disorders, developmental delays, and oppositional defiant disorder in minority African Americans.
Blood-derived whole genome sequencing data were generated from 4,179 individuals, including 1,384 patients diagnosed with at least one mental disorder. The loading of genomic variants into coding/non-coding regions was applied as feature vectors in the deep learning algorithm. Our model showed ~65% accuracy in differentiating patients from controls. The ability to label patients with multiple disorders was similarly successful, with a Hamming loss score of less than 0.3, while exact diagnostic matches hover around 10%. Genes in the genomic regions with the highest weights showed enrichment of biological pathways involved in immune responses, antigen/nucleic acid binding, chemokine signaling pathway, and G-protein receptor activities.
Remarkably, variants in noncoding regions (eg, ncRNA, intronic, and intergenic) performed as well as variants in coding regions; however, unlike variants in the coding region, variants in the non-coding regions do not express genomic breakpoints, while having much narrower standard deviations, indicating that they likely serve as surrogate markers.
