A model using imaging genetics analysis was able to predict and explain the degree of depression in Parkinson disease (PD) with a lower error and higher correlation than other models over a 5-fold cross-validation, according to the results of a study published in PLoS One.

Ji Hye Won, a PhD student from the Department of Electrical and Computer Engineering, Sungkyunkwan University and the Center for Neuroscience Imaging Research, Institute for Basic Science, Suwon, Korea, and colleagues conducted a retrospective analysis of de-identified data. They used diffusion MRI, T1-weighted MRI, and DNA genotyping data obtained from the Parkinson’s Progression Markers Initiative database for 81 patients with PD. Researchers obtained DNA samples genotyped by NeuroX genotyping arrays from the Parkinson’s Progression Markers Initiative and used the least absolute shrinkage and selection operator (LASSO) algorithm to identify regional imaging features that could characterize depression in Parkinson disease. They assessed depression clinically, using the geriatric depression scale.