Scientists at the National Institutes of Health have identified new genetic risk factors for two types of non-Alzheimer’s dementia. These findings were published in Cell Genomics and detail how researchers identified large-scale DNA changes, known as structural variants, by analyzing thousands of DNA samples. The team discovered several structural variants that could be risk factors Lewy body dementia (LBD) and frontotemporal dementia (FTD). The project was a collaborative effort between scientists at the National Institute of Neurological Disorders and Stroke (NINDS) and the National Institute on Aging (NIA) at NIH.
By combining computer algorithms capable of mapping structural variations across the whole genome with machine learning, the research team analyzed whole-genome data from thousands of patient samples and several thousand unaffected controls.
A previously unknown variant in the gene TCPN1 was found in samples from patients with LBD, a disease, that like Parkinson’s disease, is associated with abnormal deposits of the protein alpha-synuclein in the brain. This variant, in which more than 300 nucleotides are deleted from the gene, is associated with a higher risk for developing LBD. While this finding is new for LBD, TCPN1 is a known risk factor for Alzheimer’s disease, which could mean that this structural variant plays a role in the broader dementia population.
By looking at a group of 50 genes implicated in inherited neurodegenerative diseases, the investigators were able to identify additional rare structural variants, including several that are known to cause disease. The analyses also identified two well-established risk factors for FTD changes in the C9orf72 and MAPT genes. These proof-of-concept findings bolstered the strength of the study’s new findings by demonstrating that the algorithms were properly working.
Because reference maps for currently available structural variants are limited, the researchers generated a catalog based on the data obtained in these analyses. The analysis code and all the raw data are now available to the scientific community for use in their studies. An interactive app also allows investigators to study their genes of interest and ask which variants are present in controls vs. LBD or FTD cases. The authors assert these resources may make complex genetic data more accessible to non-bioinformatics experts, which will accelerate the pace of discovery.
