Deep joint learning diagnosis of Alzheimer’s disease based on multimodal feature fusion DOI Creative Commons
Jingru Wang,

S. P. Wen,

Wenjie Liu

et al.

BioData Mining, Journal Year: 2024, Volume and Issue: 17(1)

Published: Nov. 5, 2024

Alzheimer's disease (AD) is an advanced and incurable neurodegenerative disease. Genetic variations are intrinsic etiological factors contributing to the abnormal expression of brain function structure in AD patients. A new multimodal feature fusion called "magnetic resonance imaging (MRI)-p value" was proposed construct 3D images by introducing genes as a priori knowledge. Moreover, deep joint learning diagnostic model constructed fully learn features. One branch trained residual network (ResNet) features local pathological regions. The other learned position information regions with different changes categories subjects' brains attention convolution, then obtained discriminative probability from locations via convolution global average pooling. two branches were linearly interacted acquire basis for classifying subjects. diagnoses health control (HC), mild cognitive impairment (MCI), HC MCI performed data Disease Neuroimaging Initiative (ADNI). results showed that method achieved optimal AD-related diagnosis. classification accuracy (ACC) area under curve (AUC) three experimental groups 93.44% 96.67%, 89.06% 92%, 84% 81.84%, respectively. total six novel found be significantly associated AD, namely NTM, MAML2, NAALADL2, FHIT, TMEM132D PCSK5, which provided targets potential treatment diseases.

Language: Английский

Whole genome‐wide sequence analysis of long‐lived families (Long‐Life Family Study) identifies MTUS2 gene associated with late‐onset Alzheimer's disease DOI Creative Commons
Laura Xicota, Stephanie Cosentino, Badri N. Vardarajan

et al.

Alzheimer s & Dementia, Journal Year: 2024, Volume and Issue: 20(4), P. 2670 - 2679

Published: Feb. 21, 2024

Late-onset Alzheimer's disease (LOAD) has a strong genetic component. Participants in Long-Life Family Study (LLFS) exhibit delayed onset of dementia, offering unique opportunity to investigate LOAD genetics.

Language: Английский

Citations

5
Identification of key genes and signaling pathway in the pathogenesis of Huntington's disease via bioinformatics and next generation sequencing data analysis DOI Creative Commons
Basavaraj Vastrad, Chanabasayya Vastrad

Egyptian Journal of Medical Human Genetics, Journal Year: 2025, Volume and Issue: 26(1)

Published: March 4, 2025

Abstract Background Huntington's disease (HD) could cause progressive motor deficits, psychiatric symptoms, and cognitive impairment. With the increasing use of pharmacotherapies theoretically target neurotransmitters, incidence HD is still not decreasing. However, molecular pathogenesis have been illuminate. It momentous to further examine HD. Methods The next generation sequencing dataset GSE105041 was downloaded from Gene Expression Omnibus (GEO) database. Using DESeq2 in R bioconductor package screen differentially expressed genes (DEGs) between samples normal control samples. ontology (GO) term REACTOME pathway enrichment were performed on DEGs. Meanwhile, using Integrated Interactions Database (IID) database Cytoscape software construct protein–protein interaction (PPI) network module analysis, identify hub with highest value node degree, betweenness, stress closeness scores. miRNA-hub gene regulatory TF-hub constructed analyzed. Receiver operating characteristic curves analysis for diagnostic genes. Results We identified 958 DEGs, consisting 479 up regulated DEGs down GO terms analyses by g:Profiler online results revealed that mainly enriched multicellular organismal process, developmental signaling GPCR MHC class II antigen presentation. Network Analyzer plugin PPI network, LRRK2, MTUS2, HOXA1, IL7R, ERBB3, EGFR, TEX101, WDR76, NEDD4L COMT selected as Hsa-mir-1292-5p, hsa-mir-4521, ESRRB SREBF1 are potential biomarkers predicted be associated Conclusion This study investigated key pathways interactions its complications, which might help reveal correlation complications. current investigation captured prediction, follow-up biological experiments enforced validation.

Language: Английский

Citations

0
The Alzheimer's Biomarker Consortium–Down Syndrome (ABC‐DS): A 10‐year report DOI Creative Commons
Benjamin L. Handen, Mark Mapstone, Sigan L. Hartley

et al.

Alzheimer s & Dementia, Journal Year: 2025, Volume and Issue: 21(5)

Published: May 1, 2025

Abstract INTRODUCTION Virtually all adults with Down syndrome (DS) will accumulate the neuropathologies associated Alzheimer's disease (AD) by age 40, majority having a clinical dementia diagnosis their middle 50s. METHODS This paper complements 2020 publication describing Biomarker Consortium–Down Syndrome (ABC‐DS) methodology highlighting protocol changes since initial funding in 2015. It describes available clinical, neuropsychological, neuroimaging, and biofluid data bio‐specimen repository. Ten years of accomplishments are summarized. RESULTS Over 500 DS 59 sibling controls have been enrolled 2015 nearly 800 follow‐up visits. More than 900 magnetic resonance imaging (MRI), amyloid positron emission tomography (PET), 600 tau PET scans conducted; multiple omics generated using over 1100 blood 100 cerebrospinal fluid (CSF) samples. DISCUSSION ABC‐DS is largest U.S.‐based, multi‐site (including United Kingdom Puerto Rico), longitudinal biomarker initiative to target at risk for AD. Highlights The Consortium—Down entering its 10th year. enrolled. conducted. Multiple positioned continue make substantial contributions field.

Language: Английский

Citations

0
Bayesian Longitudinal Network Regression With Application to Brain Connectome Genetics DOI
Chenxi Li,

Xinyuan Tian,

Shangbing Gao

et al.

Statistics in Medicine, Journal Year: 2025, Volume and Issue: 44(8-9)

Published: April 1, 2025

ABSTRACT The increasing availability of large‐scale brain imaging genetics studies enables more comprehensive exploration the genetic underpinnings functional organizations. However, fundamental analytical challenges arise when considering complex network topology connectivity, influenced by contributions and sample relatedness, particularly in longitudinal studies. In this paper, we propose a novel method named Bayesian Longitudinal Network‐Variant Regression (BLNR), which models association between variants connectivity. BLNR fills gap existing genome‐wide that primarily focus on univariate or multivariate phenotypes. Our approach jointly biological architecture connectivity associated mixed‐effect components within framework. By employing plausible prior settings posterior inference, identification significant signals their sub‐network components, providing robust inference. We demonstrate superiority our model through extensive simulations apply it to Adolescent Brain Cognitive Development (ABCD) study. This application highlights BLNR's ability estimate effects changes configurations during neurodevelopment, demonstrating its potential extend other similar problems involving relatedness network‐variate outcomes.

Language: Английский

Citations

0
Deep joint learning diagnosis of Alzheimer’s disease based on multimodal feature fusion DOI Creative Commons
Jingru Wang,

S. P. Wen,

Wenjie Liu

et al.

BioData Mining, Journal Year: 2024, Volume and Issue: 17(1)

Published: Nov. 5, 2024

Alzheimer's disease (AD) is an advanced and incurable neurodegenerative disease. Genetic variations are intrinsic etiological factors contributing to the abnormal expression of brain function structure in AD patients. A new multimodal feature fusion called "magnetic resonance imaging (MRI)-p value" was proposed construct 3D images by introducing genes as a priori knowledge. Moreover, deep joint learning diagnostic model constructed fully learn features. One branch trained residual network (ResNet) features local pathological regions. The other learned position information regions with different changes categories subjects' brains attention convolution, then obtained discriminative probability from locations via convolution global average pooling. two branches were linearly interacted acquire basis for classifying subjects. diagnoses health control (HC), mild cognitive impairment (MCI), HC MCI performed data Disease Neuroimaging Initiative (ADNI). results showed that method achieved optimal AD-related diagnosis. classification accuracy (ACC) area under curve (AUC) three experimental groups 93.44% 96.67%, 89.06% 92%, 84% 81.84%, respectively. total six novel found be significantly associated AD, namely NTM, MAML2, NAALADL2, FHIT, TMEM132D PCSK5, which provided targets potential treatment diseases.

Language: Английский

Citations

1