Innovative Machine Learning Program Revealed

Alzheimer’s disease (AD) is a complex neurodegenerative disease with genetic and environmental origins. Women experience faster cognitive decline and brain atrophy than men, and men have a higher mortality rate. Researchers at Baylor College of Medicine and the Jan Duncan Neurological Institute (Duncan NRI) at Texas Children’s Hospital used a new machine learning technique they developed, called evolutionary behavioral machine learning (EAML), to predict sexual behavior. A specific gene and a sex-specific gene were found. Molecular pathways contributing to the development and progression of this condition. This research Nature Communications.

“We have developed a proprietary machine learning software that uses a highly computational predictive index called the Evolutionary Action (EA) Score as a function of identifying genetic factors that influence Alzheimer’s disease risk in men and women. I did,” said Dr. Olivier Richtage. said a professor of biochemistry and molecular biology at Baylor College of Medicine. “This approach has allowed us to efficiently leverage large amounts of evolutionary data, allowing us to more precisely interrogate small cohorts and identify genes involved in sex-specific differences in Alzheimer’s disease. ”

EAML is an ensemble computational approach that includes nine machine learning algorithms for analyzing the functional impact of non-synonymous coding variants, defined as DNA mutations that affect the structure and function of the resulting protein, and evolutionary Estimate adverse effects on biological processes using methods. Action (EA) score.

Lichtarge and team used EAML to analyze coding variants in 2,729 AD patients and 2,441 control subjects, identifying 98 genes associated with AD. These include several genes known to play key roles in the biology of Alzheimer’s disease, and we are working to identify genes and pathways associated with complex diseases such as Alzheimer’s disease. It supports the general value of combining machine learning approaches with phylogenetic evolutionary information embedded in EAs. They also found that these genes were aberrantly expressed in AD brains, showing that they are functionally related. Specific pathways include mediating pathways of neuroinflammation, microglia and astrocyte biology, which is consistent with their potential involvement in the pathophysiology of Alzheimer’s disease.

Next, they worked with Dr. Ismael Al-Ramahi and Dr. Juan Botas., A team from the Center for Alzheimer’s and Neurodegenerative Diseases and Duncan NRI tested homologs of 98 EAML candidate genes using two Drosophila AD models. For this, they used a robot-assisted, state-of-the-art behavioral testing platform that allows for high-throughput screening. in vivo. They found that 36 genes regulate tau-induced degeneration and 29 genes regulate her Aβ42-induced neurodegeneration. These included nine genes that could ameliorate neurodegeneration caused by both tau and Aβ42, two proteins known to accumulate in AD patients. This strongly validated that the identified candidates are functionally involved in mediating neurodegeneration. in vivo He then highlighted the potential therapeutic avenues of targeting these genes.

As the aim of this study was to understand how AD is differentially expressed and progressed in males and females, we then applied EAML analysis separately to males and females within this cohort. They found 157 AD-related genes in males and 127 AD-related genes in females. Genes identified in this gender-disaggregated study were found to be more closely related to known AD GWAS genes than genes identified in mixed-gender studies. These findings suggest that gender analysis increases the sensitivity of identification of Alzheimer’s disease-associated genes and improves risk prediction ability.

Furthermore, they found that certain biological pathways may have a more significant impact on the development of Alzheimer’s disease in one sex than the other. For example, female-specific EAML candidates were found to be involved in modules related to cell cycle control and DNA quality control. “We are thrilled to have discovered a gene cluster associated with BRCA1, a gene known to be neuroprotective in women and associated with breast cancer. The two diseases suggest a biological relevance that may occur more often in women than in men,” said Dr. Ismael Al-Ramahi. These findings may have important implications for the development of therapeutic strategies and the design of gender-stratified clinical trials for Alzheimer’s disease.

Additionally, EAML maintained consistent and robust target-driven predictive capabilities even when the team tested with smaller sample sizes. With just 700 samples, EAML was able to recover more than 50% of the candidates found across the dataset, significantly outperforming prediction algorithms currently in use. The authors conclude that this greatly improved capability will enable researchers to reach accurate and reliable predictions using smaller datasets, where known methods yield reliable results. We believe this will pave the way for incorporating gender-specific analysis into disease-gene association studies that may not have been possible.

“The success of using EAML to discover new targets for Alzheimer’s disease not only provides new perspectives on the genetic factors that influence this disease, but also explores disease-gene associations. It also underscores the importance of systematically applying gender-specific analysis when conducting research,” said Dr. Juan Botas. added the professor of molecular and human genetics at Baylor University. “This revolutionary approach has the potential to revolutionize our understanding of complex diseases such as Alzheimer’s disease and drive the development of personalized therapies tailored to each individual’s genetic makeup.”

Other members involved in the study include Thomas Bulkar, Kwanhyuk Lee, Min Pham, Dillon Shapiro, Yashwanth Ragissetti, Shirin Soleimani, Samantha Mota and Kevin Wilhelm. , Mariam Saminasab, Yongwon Kim, Unna Fu, Jennifer Asmussen and Panagiotis Katssonis. They are affiliated with his one or more of the following institutions: Baylor College of Medicine, The Jan and Dunduncan Neurological Institute at Texas Children’s Hospital, UTHealth McGovern School of Medicine. This study was funded by a grant from the National Institutes of Health.

Disclaimer: AAAS and EurekAlert! We are not responsible for the accuracy of news releases posted on EurekAlert!. Use of information by contributing institutions or via the EurekAlert system.