{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T16:07:26Z","timestamp":1781626046317,"version":"3.54.5"},"reference-count":41,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2025,12,22]],"date-time":"2025-12-22T00:00:00Z","timestamp":1766361600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,12,29]],"date-time":"2025-12-29T00:00:00Z","timestamp":1766966400000},"content-version":"vor","delay-in-days":7,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BioData Mining"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Alzheimer\u2019s Disease (AD) represents a growing global health challenge, driven by complex genetic factors and diverse risk contributors. Currently, an estimated 55 million people worldwide are affected by dementia, with AD responsible for 60\u201370% of these cases. This paper explores the application of advanced machine learning approaches to predict AD risk using Genome-Wide Association Studies data from multiple cohorts, with a particular focus on transfer learning and feature selection techniques. We evaluate the performance of Wide and Deep Neural Networks and Multi-Head Attention in assessing their ability to generalise across datasets. As part of this, we explore knowledge distillation as a strategy to enhance model efficiency through improved generalisation performance in smaller architectures by transferring knowledge from high-capacity models to lightweight ones. Furthermore, the performance of these deep learning approaches is compared with tree-based ensembles, including Random Forest and XGBoost. Our experiments evaluate the generalisability, transferability, and efficiency of these models across different transfer learning scenarios. Findings indicate that aggregating multi-cohort training data significantly enhances predictive performance, highlighting the importance of data diversity in improving AD risk assessment. The proposed knowledge distillation approach enables the transfer of knowledge from a complex teacher model to a simpler student model, significantly improving performance. To enhance interpretability, we apply SHAP (SHapley Additive exPlanations) to the student models, revealing cohort-specific differences in SNP importance and highlighting variants in genes such as ABI3BP and SYN3, both of which are linked to immune and synaptic functions in AD. The integration of SHAP enables transparent interpretation of model decisions and supports the identification of transferable genetic markers, reinforcing the clinical relevance of our framework in AD risk prediction.<\/jats:p>","DOI":"10.1186\/s13040-025-00506-0","type":"journal-article","created":{"date-parts":[[2025,12,22]],"date-time":"2025-12-22T10:38:33Z","timestamp":1766399913000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Cross-cohort genetic risk prediction for Alzheimer\u2019s disease: a transfer learning approach using GWAS and deep learning models"],"prefix":"10.1186","volume":"18","author":[{"given":"Isibor Kennedy","family":"Ihianle","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wathsala","family":"Samarasekara","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Keeley","family":"Brookes","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pedro","family":"Machado","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2025,12,22]]},"reference":[{"issue":"24","key":"506_CR1","doi-asserted-by":"publisher","first-page":"5789","DOI":"10.3390\/molecules25245789","volume":"25","author":"Z Breijyeh","year":"2020","unstructured":"Breijyeh Z, Karaman R. Comprehensive review on alzheimer\u2019s disease: causes and treatment. Molecules. 2020;25(24):5789.","journal-title":"Molecules"},{"key":"506_CR2","doi-asserted-by":"crossref","unstructured":"Kirshner HS. Memory loss, Alzheimer\u2019s disease, and dementia: a practical guide for clinicians. Cogn Behav Neurol. 2022;35(4):298\u20139.","DOI":"10.1097\/WNN.0000000000000323"},{"key":"506_CR3","unstructured":"Organization WH. Dementia. 2024. https:\/\/www.who.int\/news-room\/fact-sheets\/detail\/dementia. Accessed 6 Feb 2024."},{"key":"506_CR4","unstructured":"International AD. Dementia statistics. 2024. https:\/\/www.alzint.org\/about\/dementia-facts-figures\/dementia-statistics\/. Accessed 6 Feb 2024."},{"key":"506_CR5","doi-asserted-by":"crossref","unstructured":"Saragea PD. The genetic landscape of early and late-onset alzheimer\u2019s disease-a review. Acta Marisiensis-Seria Medica. 2024;70(4).","DOI":"10.2478\/amma-2024-0030"},{"issue":"6","key":"506_CR6","doi-asserted-by":"publisher","first-page":"110","DOI":"10.3390\/diseases12060110","volume":"12","author":"CA Valdez-Gaxiola","year":"2024","unstructured":"Valdez-Gaxiola CA, Rosales-Leycegui F, Gaxiola-Rubio A, Moreno-Ortiz JM, Figuera LE. Early-and late-onset alzheimer\u2019s disease: two sides of the same coin? Diseases. 2024;12(6):110.","journal-title":"Diseases"},{"issue":"18","key":"506_CR7","doi-asserted-by":"publisher","first-page":"2272","DOI":"10.1212\/WNL.0000000000011772","volume":"96","author":"DS Smirnov","year":"2021","unstructured":"Smirnov DS, Galasko D, Hiniker A, Edland SD, Salmon DP. Age-at-onset and apoe-related heterogeneity in pathologically confirmed sporadic alzheimer disease. Neurology. 2021;96(18):2272\u201383.","journal-title":"Neurology"},{"key":"506_CR8","doi-asserted-by":"publisher","first-page":"2594","DOI":"10.1007\/s11682-019-00212-6","volume":"14","author":"L Pini","year":"2020","unstructured":"Pini L, Geroldi C, Galluzzi S, Baruzzi R, Bertocchi M, Chito E, Orini S, Romano M, Cotelli M, Rosini S, et al. Age at onset reveals different functional connectivity abnormalities in prodromal alzheimer\u2019s disease. Brain Imag Behav. 2020;14:2594\u2013605.","journal-title":"Brain Imag and Behav"},{"issue":"7","key":"506_CR9","doi-asserted-by":"publisher","first-page":"849","DOI":"10.1001\/jamaneurol.2020.0414","volume":"77","author":"ME Belloy","year":"2020","unstructured":"Belloy ME, Napolioni V, Han SS, Le Guen Y, Greicius MD, Initiative ADN, et al. Association of klotho-vs heterozygosity with risk of alzheimer disease in individuals who carry apoe4. JAMA Neurol. 2020;77(7):849\u201362.","journal-title":"JAMA Neurol"},{"key":"506_CR10","doi-asserted-by":"crossref","unstructured":"Hettiarachchi G, Komar AA. Gwas to identify snps associated with common diseases and individual risk: genome wide association studies (gwas) to identify snps associated with common diseases and individual risk. Single Nucleotide Polymorphisms: Hum Variation a Coming Revolution Biol Med. 2022;51\u201376.","DOI":"10.1007\/978-3-031-05616-1_4"},{"issue":"10","key":"506_CR11","doi-asserted-by":"publisher","first-page":"5797","DOI":"10.1038\/s41380-021-01152-8","volume":"26","author":"MAA DeMichele-Sweet","year":"2021","unstructured":"DeMichele-Sweet MAA, Klei L, Creese B, Harwood JC, Weamer EA, McClain L, Sims R, Hernandez I, Moreno-Grau S, T\u00e1rraga L, et al. Genome-wide association identifies the first risk loci for psychosis in alzheimer disease. Mol Psychiatry. 2021;26(10):5797\u2013811.","journal-title":"Mol Psychiatry"},{"issue":"2","key":"506_CR12","doi-asserted-by":"publisher","first-page":"359","DOI":"10.31083\/j.jin.2020.02.110","volume":"19","author":"M Prendecki","year":"2020","unstructured":"Prendecki M, Kowalska M, Lagan-Jkedrzejczyk U, Piekut T, Krokos A, Kozubski W, Dorszewska J. Genetic factors related to the immune system in subjects at risk of developing alzheimer\u2019s disease. J Intgr Neurosci. 2020;19(2):359\u201371.","journal-title":"J Intgr Neurosci"},{"key":"506_CR13","doi-asserted-by":"crossref","unstructured":"Li Y, Laws SM, Miles LA, Wiley JS, Huang X, Masters CL, Gu BJ. Genomics of alzheimer\u2019s disease implicates the innate and adaptive immune systems. Cellular Mol Life Sci. 2021;1\u201330.","DOI":"10.1007\/s00018-021-03986-5"},{"key":"506_CR14","doi-asserted-by":"crossref","unstructured":"Hwang H, Bell A, Fonseca J, Pliatsika V, Stoyanovich J, Whang SE. Shap-based explanations are sensitive to feature representation. Proceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency. 2025, pp. 1588\u2013601.","DOI":"10.1145\/3715275.3732105"},{"key":"506_CR15","doi-asserted-by":"crossref","unstructured":"Lambert JC, A. Meta-analysis of 74,046 individuals identifies 11 new susceptibility loci for alzheimer\u2019s disease. Nat Genet. 2013;45(12):1452\u201358.","DOI":"10.1038\/ng.2802"},{"issue":"3","key":"506_CR16","doi-asserted-by":"publisher","first-page":"414","DOI":"10.1038\/s41588-019-0358-2","volume":"51","author":"BW Kunkle","year":"2019","unstructured":"Kunkle BW, Grenier-Boley B, Sims R, Bis JC, Damotte V, Naj AC, Boland A, Vronskaya M, Van Der Lee SJ, Amlie-Wolf A, et al. Genetic meta-analysis of diagnosed alzheimer\u2019s disease identifies new risk loci and implicates a\u03b2, tau, immunity and lipid processing. Nat Genet. 2019;51(3):414\u201330.","journal-title":"Nat Genet"},{"issue":"3","key":"506_CR17","doi-asserted-by":"publisher","first-page":"404","DOI":"10.1038\/s41588-018-0311-9","volume":"51","author":"IE Jansen","year":"2019","unstructured":"Jansen IE, Savage JE, Watanabe K, Bryois J, Williams D, Steinberg S, Sealock J, Karlsson IK, H\u00e4gg S, Athanasiu L, et al. Genome-wide meta-analysis identifies new loci and functional pathways influencing alzheimer\u2019s disease risk. Nat Genet. 2019;51(3):404\u201313.","journal-title":"Nat Genet"},{"issue":"3","key":"506_CR18","doi-asserted-by":"publisher","first-page":"589","DOI":"10.1016\/j.cell.2019.08.051","volume":"179","author":"RE Peterson","year":"2019","unstructured":"Peterson RE, Kuchenbaecker K, Walters RK, Chen CY, Popejoy AB, Periyasamy S, Lam M, Iyegbe C, Strawbridge RJ, Brick L, Carey CE, Martin AR, Meyers JL, Su J, Chen J, Edwards AC, Kalungi A, Koen N, Majara L, Schwarz E, Duncan LE. Genome-wide association studies in ancestrally diverse populations: opportunities, methods, pitfalls, and recommendations. Cell. 2019;179(3):589\u2013603. https:\/\/doi.org\/10.1016\/j.cell.2019.08.051.","journal-title":"Cell"},{"issue":"3","key":"506_CR19","doi-asserted-by":"publisher","first-page":"265","DOI":"10.1037\/neu0000720","volume":"35","author":"LB Zahodne","year":"2021","unstructured":"Zahodne LB, Sharifian N, Kraal AZ, Zaheed AB, Sol K, Morris EP, Schupf N, Manly JJ, Brickman AM. Socioeconomic and psychosocial mechanisms underlying racial\/ethnic disparities in cognition among older adults. Neuropsychology. 2021;35(3):265.","journal-title":"Neuropsychology"},{"key":"506_CR20","doi-asserted-by":"publisher","first-page":"1227","DOI":"10.3389\/fgene.2019.01227","volume":"10","author":"JM Luningham","year":"2019","unstructured":"Luningham JM, McArtor DB, Hendriks AM, Beijsterveldt CE, Lichtenstein P, Lundstr\u00f6m S, Larsson H, Bartels M, Boomsma DI, Lubke GH. Data integration methods for phenotype harmonization in multi-cohort genome-wide association studies with behavioral outcomes. Front Genet. 2019;10:1227.","journal-title":"Front Genet"},{"key":"506_CR21","doi-asserted-by":"publisher","unstructured":"Pudjihartono N, Fadason T, Kempa-Liehr A, O. J. A review of feature selection methods for machine learning-based disease risk prediction. Front Bioinf. 2022;2:927312. https:\/\/doi.org\/10.3389\/fbinf.2022.927312.","DOI":"10.3389\/fbinf.2022.927312"},{"key":"506_CR22","doi-asserted-by":"publisher","unstructured":"Alatrany AS, Khan W, Hussain A, Al-Jumeily D. Alzheimer\u2019s disease neuroimaging initiative: wide and deep learning based approaches for classification of alzheimer\u2019s disease using genome-wide association studies. PLoS One. 2023;18(5):0283712 https:\/\/doi.org\/10.1371\/journal.pone.0283712.","DOI":"10.1371\/journal.pone.0283712"},{"key":"506_CR23","doi-asserted-by":"publisher","unstructured":"Alshamlan H, Omar S, Aljurayyad R, Alabduljabbar R. Identifying effective feature selection methods for alzheimer\u2019s disease biomarker gene detection using machine learning. Diagnostics. 2023;13(10). https:\/\/doi.org\/10.3390\/diagnostics13101771.","DOI":"10.3390\/diagnostics13101771"},{"issue":"2","key":"506_CR24","doi-asserted-by":"publisher","first-page":"998","DOI":"10.1016\/j.eswa.2009.05.075","volume":"37","author":"S G\u00fcne\u015f","year":"2010","unstructured":"G\u00fcne\u015f S, Polat K, Yosunkaya. Multi-class f-score feature selection approach to classification of obstructive sleep apnea syndrome. Expert Syst Appl. 2010;37(2):998\u20131004. https:\/\/doi.org\/10.1016\/j.eswa.2009.05.075.","journal-title":"Expert Syst With Appl"},{"issue":"10","key":"506_CR25","doi-asserted-by":"publisher","first-page":"1345","DOI":"10.1109\/TKDE.2009.191","volume":"22","author":"SJ Pan","year":"2010","unstructured":"Pan SJ, Yang Q. A survey on transfer learning. IEEE Trans Knowl Data Eng. 2010;22(10):1345\u201359.","journal-title":"IEEE Trans On Knowl And Data Eng"},{"key":"506_CR26","unstructured":"Ruder S. An overview of multi-task learning in deep neural networks. arXiv:1706.05098."},{"key":"506_CR27","unstructured":"Wang Z, A. Transfer learning for gwas: applications in alzheimer\u2019s disease risk prediction. Genet in Med. 2021;23(8):1404\u201312."},{"key":"506_CR28","doi-asserted-by":"crossref","unstructured":"Abbasi S, Hajabdollahi M, Karimi N, Samavi S. Modeling teacher-student techniques in deep neural networks for knowledge distillation. In: 2020 International Conference on Machine Vision and Image Processing (MVIP). IEEE; 2020. p. 1\u20136.","DOI":"10.1109\/MVIP49855.2020.9116923"},{"key":"506_CR29","doi-asserted-by":"crossref","unstructured":"Waheed Z, Gui J, Heyat MBB, Parveen S, Hayat MAB, Iqbal MS, Aya Z, Nawabi AK, Sawan M. A novel lightweight deep learning based approaches for the automatic diagnosis of gastrointestinal disease using image processing and knowledge distillation techniques. Comput Methods Programs Biomed. 2025;260:108579.","DOI":"10.1016\/j.cmpb.2024.108579"},{"key":"506_CR30","unstructured":"Balboni D, Bacciu D. A partial theory of wide neural networks using wc functions and its practical implications."},{"key":"506_CR31","doi-asserted-by":"publisher","first-page":"159","DOI":"10.1016\/j.neurobiolaging.2021.05.014","volume":"107","author":"J Young","year":"2021","unstructured":"Young J, Gallagher E, Koska K, Guetta-Baranes T, Morgan K, Thomas A, Brookes KJ. Genome-wide association findings from the brains for dementia research cohort. Neurobiol of Aging. 2021;107:159\u201367.","journal-title":"Neurobiol of Aging"},{"issue":"1","key":"506_CR32","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/sdata.2018.142","volume":"5","author":"PL De Jager","year":"2018","unstructured":"De Jager PL, Ma Y, McCabe C, Xu J, Vardarajan BN, Felsky D, Klein H-U, White CC, Peters MA, Lodgson B, et al. A multi-omic atlas of the human frontal cortex for aging and alzheimer\u2019s disease research. Sci Data. 2018;5(1):1\u201313.","journal-title":"Sci Data"},{"key":"506_CR33","doi-asserted-by":"publisher","unstructured":"Sevani N, Hermawan I, Jatmiko W. Feature selection based on f-score for enhancing ctg data classification. 2019 IEEE International Conference on Cybernetics and Computational Intelligence (CyberneticsCom). 2019:18\u201322. https:\/\/doi.org\/10.1109\/CYBERNETICSCOM.2019.8875656.","DOI":"10.1109\/CYBERNETICSCOM.2019.8875656"},{"key":"506_CR34","doi-asserted-by":"crossref","unstructured":"Jalali-Najafabadi F, Stadler M, Dand N, Jadon D, Soomro M, Ho P, Marzo-Ortega H, Helliwell P, Korendowych E, Simpson MA, et al. Application of information theoretic feature selection and machine learning methods for the development of genetic risk prediction models. Sci Rep. 2021;11(1):23335.","DOI":"10.1038\/s41598-021-00854-x"},{"key":"506_CR35","doi-asserted-by":"crossref","unstructured":"Ihianle IK, Machado P, Owa K, Adama DA, Otuka R, Lotfi A. Minimising redundancy, maximising relevance: HRV feature selection for stress classification. Expert Syst Appl. 2024;239:122490.","DOI":"10.1016\/j.eswa.2023.122490"},{"key":"506_CR36","unstructured":"Tang J, Shivanna R, Zhao Z, Lin D, Singh A, Chi EH, Jain S. Understanding and improving knowledge distillation. arXiv preprint arXiv:2002.03532. 2020."},{"key":"506_CR37","doi-asserted-by":"publisher","first-page":"709","DOI":"10.1186\/s12859-019-3158-x","volume":"20","author":"JDV Oriol","year":"2019","unstructured":"Oriol JDV, Vallejo EE, Estrada K, Pe\u00f1a JGT, Initiative ADN, et al. Benchmarking machine learning models for late-onset alzheimer\u2019s disease prediction from genomic data. BMC Bioinf. 2019;20:709.","journal-title":"BMC Bioinf"},{"key":"506_CR38","doi-asserted-by":"crossref","unstructured":"Arnal Segura M, Bini G, Fernandez Orth D, Samaras E, Kassis M, Aisopos F, et al. Machine learning methods applied to genotyping data capture interactions between single nucleotide variants in late onset Alzheimer\u2019s disease. Alzheimers Dement (Diagn Assess Dis Monit). 2022;14(1):12300.","DOI":"10.1002\/dad2.12300"},{"issue":"2","key":"506_CR39","doi-asserted-by":"publisher","first-page":"022","DOI":"10.1093\/bib\/bbac022","volume":"23","author":"T Jo","year":"2022","unstructured":"Jo T, Nho K, Bice P, Saykin AJ, Initiative, A.D.N. Deep learning-based identification of genetic variants: application to alzheimer\u2019s disease classification. Briefings Bioinf. 2022;23(2):022.","journal-title":"Briefings Bioinf"},{"issue":"1","key":"506_CR40","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1093\/jlb\/lsz007","volume":"6","author":"EW Clayton","year":"2019","unstructured":"Clayton EW, Evans BJ, Hazel JW, Rothstein MA. The law of genetic privacy: applications, implications, and limitations. J Law Biosci. 2019;6(1):1\u201336.","journal-title":"J Law Biosci"},{"key":"506_CR41","first-page":"191","volume":"104","author":"A Act","year":"1996","unstructured":"Act A. Health insurance portability and accountability act of 1996. Public Law. 1996;104:191.","journal-title":"Public Law"}],"container-title":["BioData Mining"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s13040-025-00506-0","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13040-025-00506-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13040-025-00506-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T15:34:19Z","timestamp":1781624059000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1186\/s13040-025-00506-0"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,22]]},"references-count":41,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2025,12]]}},"alternative-id":["506"],"URL":"https:\/\/doi.org\/10.1186\/s13040-025-00506-0","relation":{},"ISSN":["1756-0381"],"issn-type":[{"value":"1756-0381","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12,22]]},"assertion":[{"value":"22 April 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"29 November 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"22 December 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"This study utilised publicly available genomic datasets and ethically approved cohort resources. The Brains for Dementia Research (BDR) cohort used in this study has ethical approval from the London \u2013 City and East NRES Committee (Reference: 08\/H0704\/128+5). All tissue requests approved under this programme are covered by this ethical clearance. The BDR initiative is jointly funded by Alzheimer\u2019s Research UK and the Alzheimer\u2019s Society in partnership with the Medical Research Council. Genotypic data from the Alzheimer\u2019s Disease Centres seventh set of ADC genotyped subjects (ADC7; NG00071), the National Institute on Aging Genetics Initiative for Late-Onset Alzheimer\u2019s Disease (NIA-LOAD; NG00020), and TGen II (TGEN; NG00028) were obtained through the NIAGADS data repository and used in accordance with their data usage agreements. The BDR genotypes are available via the Dementia Platform UK. All analyses in this study adhered to ethical guidelines for the use of anonymised human genomic data and complied with relevant institutional, national, and international regulations, including the Declaration of Helsinki. All participants (or their legal representatives) provided informed consent for the use of their data, or data were fully anonymised in accordance with institutional and data governance policies. No identifying information is included in the published datasets; all data are anonymised to protect participant privacy. Access to certain datasets is subject to approval by the relevant data access committees, in accordance with data usage agreements.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethical approval"}},{"value":"Not applicable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare no competing interests.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interest"}}],"article-number":"89"}}