Summary
Augmenting traditional genome wide association studies (GWAS) with advanced machine learning algorithms can allow the detection of novel signals in available cohorts. We introduce “Genome wide association neural networks (GWANN)”, a novel approach that uses neural networks (NNs) to perform a gene-level association study with family history of Alzheimer’s disease (AD). In UK Biobank, we defined cases (n=42,110) as those with AD or family history of AD and sampled an equal number of controls. The data was split into an 80:20 ratio of training and testing samples, and GWANN was trained on the former followed by identifying associated genes using its performance on the latter. Our method identified 18 genes to be associated with family history of AD. APOE, BIN1, SORL1, ADAM10, APH1B, and SPI1 have also been identified by previous AD GWAS. PCDH9, NGR3, ROR1, LINGO2, SMYD3, and LRRC7 were among the new genes that have been previously associated with neurofibrillary tangles or phosphorylated tau. Furthermore, there is evidence for differential transcriptomic or proteomic expression between AD and healthy brains for 11 of the 12 new genes. A series of post-hoc analyses resulted in a significantly enriched protein-protein interaction network (P-value<1×10−16), and enrichment of relevant disease and biological pathways such as focal adhesion (P-value=1×10−4), extracellular matrix organisation (P-value=1×10−4), Hippo signalling (P-value=7×10−4), Alzheimer’s disease (P-value=3×10−4), impaired cognition (P-value=4×10−3), and autism spectrum disorders (P-value=1×10−2). Applying NNs for GWAS illustrates their potential to complement existing algorithms and methods and enable the discovery of new associations without the need to expand existing cohorts.
Competing Interest Statement
W.S. received funding from Johnson & Johnson. A.N.H. received funding from Johnson & Johnson, GlaxoSmithKline, and Ono Pharma. T.Z. received funding from Novo Nordisk.
Funding Statement
This work was supported by the Centre for Artificial Intelligence in Precision Medicines; Johnson and Johnson; the John Fell Foundation [grant ID 0010659]; and the Virtual Brain Cloud from European commission [grant number H2020-SC1-DTH-2018-1]. C.A. is funded by the National Institute for Health Research (NIHR) Oxford Biomedical Research Centre (BRC). The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health.
Author Declarations
I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.
Yes
The details of the IRB/oversight body that provided approval or exemption for the research described are given below:
The UK Biobank gave ethical approval for this work under the approved project 15181.
I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.
Yes
I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).
Yes
I have followed all appropriate research reporting guidelines and uploaded the relevant EQUATOR Network research reporting checklist(s) and other pertinent material as supplementary files, if applicable.
Yes
Footnotes
The modifications to the analysis broadly encompass (i) changing the phenotype of interest to family history of AD instead of separately analysing maternal history and paternal history, followed by a meta-analysis; (ii) using imputed genotype dosages instead of hard-call genotypes and encoding missing values as –1; and (iii) increasing the stability of the method to reduce false positives. The modified analysis identified 18 genes associated with family history of Alzheimer's disease in the UK Biobank. Among these, 6 were well-known Alzheimer's genes, and 12 were new genes, some of which have been previously linked to neurofibrillary tangles, phosphorylated tau, or differential expression between Alzheimer's and healthy brains.
Data Availability
All data produced in the present work are contained in the manuscript and supplementary information and files.