Machine learning algorithms can be trained to estimate age from brain structural MRI. The difference between an individual s predicted and chronological age, predicted age difference (PAD), is a phenotype of relevance to aging and brain disease. In our study, we presented a new deep learning approach to predict brain age from a T1-weighted MRI. The method was trained on a dataset of healthy Icelanders and tested on two datasets, IXI and UK Biobank, utilizing transfer learning to improve accuracy on new sites. A genome-wide association study (GWAS) of PAD in the UK Biobank data (discovery set: N=12378, replication set: N=4456) yielded two sequence variants, rs1452628-T ( =-0.08, P=1.15 10-9) and rs2435204-G ( =0.102, P=9.73 10-12). The former is near KCNK2 and correlates with reduced sulcal width, whereas the latter correlates with reduced white matter surface area and tags a well-known inversion at 17q21.31 (H2).
Genetics of lumbar disc disease and associated sciatica
Aims:To replicate, in a different population, findings from a GWAS in Iceland on lumbar disc disease. For these purposes, we are requesting access to UK Biobank genotype data for relevant phenotypes and age and gender-matched population controls. The stated aim of UK Biobank is to support research that improves prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses. While lumbar disc disease is not life-threatening, it causes serious pain and disability. Indeed, chronic lower back pain is the most common cause of disability in young adults (<45); Its socioeconomic impact is therefore considerable. The pathophysiology of lumbar disc disease or its consequences is not well understood. Genetic findings can improve understanding of the biological underpinnings of this complex and common condition and generate knowledge that could eventually improve prevention and treatment. A set of common markers associating with lumbar disc disease have been discovered in the Icelandic population. To replicate our findings, we are requesting access to genotype data associated with relevant lumbar disc disease phenotypes and population control data available in the UK Biobank. We intend to conduct a case/control GWAS using the UK Biobank data and compare results with our GWAS findings in the Icelandic population. As the proposed research involves cases and controls, we would like to receive the full cohort with the requested subset of phenotype data.
|Lead investigator:||Professor Kari Stefansson|
|Lead institution:||deCODE genetics ehf|
|2283||Brain age prediction using deep learning uncovers associated sequence variants||Jonsson, B.A. et al||2019||Nature Communications. 2019|