The detection of Parkinson’s disease (PD) in its early stages is of great importance for its treatment and management, but consensus is lacking on what information is necessary and what models should be used to best predict PD risk. In our study, we first grouped PD-associated factors based on their cost and accessibility, and then gradually incorporated them into risk predictions, which were built using eight commonly used machine learning models to allow for comprehensive assessment. Finally, the Shapley Additive Explanations (SHAP) method was used to investigate the contributions of each factor. We found that models built with demographic variables, hospital admission examinations, clinical assessment, and polygenic risk score achieved the best prediction performance, and the inclusion of invasive biomarkers could not further enhance its accuracy. Among the eight machine learning models considered, penalized logistic regression and XGBoost were the most accurate algorithms for assessing PD risk, with penalized logistic regression achieving an area under the curve of 0.94 and a Brier score of 0.08. Olfactory function and polygenic risk scores were the most important predictors for PD risk. Our research has offered a practical framework for PD risk assessment, where necessary information and efficient machine learning tools were highlighted.

Type 2 diabetes (T2D) is often comorbid with cardiovascular diseases (CVDs). The direction of causation between T2D and CVD is difficult to determine; however, there may be a common underlying pathway attributable to shared genetic factors. We aimed to determine whether there is a shared genetic susceptibility to T2D and CVD. This study utilizes large-scale datasets from the UK Biobank (UKB) and DIAGRAM consortium to investigate the genetic association between T2D and CVD through phenotypic association analyses, linkage disequilibrium score (LDSC) analysis, and polygenic risk score (PRS) analysis. LDSC analysis demonstrates significant genetic associations between T2D and various CVD subtypes, including angina, heart failure (HF), myocardial infarction (MI), peripheral vascular disease (PVD), and stroke. Although the genetic association between T2D and atrial fibrillation (AF) was not significant, individuals in the high-T2D PRS group had a significantly increased risk of CVD. These findings suggest a common genetic basis and suggest that genetic susceptibility to T2D may be a potential predictor of CVD risk.

Parkinson’s disease (PD) is a complex neurodegenerative disorder with unclear etiology and ineffective treatments. Integrating multimodal data for PD prediction remains challenging. We analyzed data obtained from the Parkinson’s Progression Markers Initiative, using polygenic risk scores (PRS) to reflect genetic susceptibility to PD. We compared the prediction accuracy of models with PRS, demographics, clinical assessment, and biomarkers progressively integrated and investigated relationships. The SDPR-based PRS exhibited the highest prediction performance with an AUC of 0.75. Models combining PRS, demographic, and clinical variables achieved an AUC of 0.91, surpassing models without PRS and matching those with biomarkers. PRS correlated with olfactory function and Movement Disorder Society Unified Parkinson’s Disease Rating Scale (MDS-UPDRS), with its influence on PD risk dependent on gender and MDS-UPDRS. Our study illuminates PD etiology and provides a practical risk assessment framework, highlighting its omnigenic architecture, and the potential for accurate prediction using PRS and non-invasive clinical data.