biostudies-literatureJo TEuropean Research CouncilNIA NIH HHSNHLBI NIH HHSDutch Research Council (NWO)NHGRI NIH HHSNLM NIH HHSNINDS NIH HHSWellcome TrustNational Human Genome Research InstituteNational Institute on Aginge70041https://www.ebi.ac.uk/biostudies/studies/S-EPMC11736638biostudies-literatureUnknown11(1)<h4>Introduction</h4>The exponential growth of genomic datasets necessitates advanced analytical tools to effectively identify genetic loci from large-scale high throughput sequencing data. This study presents Deep-Block, a multi-stage deep learning framework that incorporates biological knowledge into its AI architecture to identify genetic regions as significantly associated with Alzheimer's disease (AD). The framework employs a three-stage approach: (1) genome segmentation based on linkage disequilibrium (LD) patterns, (2) selection of relevant LD blocks using sparse attention mechanisms, and (3) application of TabNet and Random Forest algorithms to quantify single nucleotide polymorphism (SNP) feature importance, thereby identifying genetic factors contributing to AD risk.<h4>Methods</Alzheimer's & dementia (New York, N. Y.)LD-informed deep learning for Alzheimer's gene loci detection using WGS data.PMC11736638RC2 HL102419U01 AG062943U19 AG024904047.017.043R01 AG009956R01 AG019085R37 AG015473U54 HG003079P20 AG068053T32 AG071444HHSN268201100009IU01 HL096902R01 AG023629P30 AG066444R01 AG048927R01 AG022018P01 NS026630HHSN268201100009CR01 AG027944R01 AG057909R01 AG064614RF1 AG054080P30 AG072958P30 AG072959RF1 AG058066P30 AG066546U54 HG003067R03 AG063250HHSN268200800007CRF1 AG058501R01 AG032289N01 HC055222RF1 AG053303P30 AG038072340755R01 AG019757RF1 AG054074RC2 HG005605U01 AG032984U24 AG041689P50 NS039764P30 AG072946P30 AG072947P01 AG003991R56 AG051876R01 AG027161U19 AG066567P30 AG010133R01 AG057739R01 AG060747U01 HL080295R01 AG068193R01 AG032990U24 AG021886P30 AG072931P20 AG068077R01 AG041797R01 AG044546R01 AG033040R01 AG020098U01 NS041588R01 AG054076R01 NS017950RF1 AG051504R01 AG044829U01 AG068057R01 AG019771R01 AG049607R01 AG011380P50 AG016574P30 AG066530P20 AG068082RF1 AG054052R01 AG048234U24 AG056270R01 AG025259R01 AG017917U01 AG058654R01 HL105756R01 AG061155U01 HL130114R01 AG054047N01 HC085083N01 HC085082N01 HC085081P30 AG062677N01 HC085086R01 NS029993U54 HG003273P50 AG008012P30 AG066514P30 AG066515U01 AG046139N01 HC085080P30 AG066518P30 AG066519P30 AG062715HHSN268201100005CP50 NS071674R01 AG046949U01 AG058922R01 HL070825HHSN268201100005IR01 AG042188HHSN268201100005GU54 AG052427R01 AG015928HHSN268201500001CP30 AG062421P30 AG062422R01 AG031272P30 AG010161RF1 AG058267U01 AG052409P30 AG066506P30 AG066507P30 AG066508P30 AG062429P30 AG066509U19 AG074879HHSN268201100006CP30 AG066511P30 AG066512U24 AG072122R01 NS080820U01 AG058635U01 AG052411P20 AG068024HHSN268201500001IHHSN268201100010CR01 AG028786U01 AG052410R01 AG015819R01 AG059716UF1 AG047133U01 AG062602175.010.200HHSN268201100011IRC4 AG039085R01 AG058066U01 AG006781U01 HL096917R01 AG018016R01 AG021547P30 AG072975U01 AG072177P30 AG066462P30 AG072976P30 AG072977U24 AG074855HHSN268201200036CRF1 AG054023HHSN268201100007CU01 AG057659P30 AG072972P30 AG072973P30 AG066468R01 AG064877P30 AG072978R01 NS069719HHSN268201100011CR01 AG079280U01 AG058589P30 AG072979HHSN268201100007IRF1 AG015473R01 AG009029R01 LM012535P50 AG005136R01 AG011101U01 HL096814HHSN268201100008CU01 HL096899R01 AG021136RF1 AG057519U01 AG049505R01 AG033193U01 AG049506U01 HL096812U01 AG049507U01 AG049508U01 AG024904P01 AG026276N01 HC085079R01 LM013463K23 AG030944R01 AG042210HHSN268201100008IHHSN268201100012CSaykin AJJo TAlzheimer's Disease Sequencing ProjectBice PNho KfalseLD-informed deep learning for Alzheimer's gene loci detection using WGS data.<h4>Introduction</h4>The exponential growth of genomic datasets necessitates advanced analytical tools to effectively identify genetic loci from large-scale high throughput sequencing data. This study presents Deep-Block, a multi-stage deep learning framework that incorporates biological knowledge into its AI architecture to identify genetic regions as significantly associated with Alzheimer's disease (AD). The framework employs a three-stage approach: (1) genome segmentation based on linkage disequilibrium (LD) patterns, (2) selection of relevant LD blocks using sparse attention mechanisms, and (3) application of TabNet and Random Forest algorithms to quantify single nucleotide polymorphism (SNP) feature importance, thereby identifying genetic factors contributing to AD risk.<h4>Methods</2025-01-01T00:00:00Z2025 Jan-Mar2026-06-03T01:16:29.212Z2025-04-06T07:42:20.82ZS-EPMC117366383982259010.1002/trc2.70041