The latest medical research paper published in the academic journal Nature Communications under Springer Nature suggests that using machine learning models to detect proteins in the blood may help predict the onset of Parkinson's disease as early as 7 years before the onset of motor symptoms. Parkinson's disease is a neurodegenerative disease characterized by delayed movement, stiffness, and static tremors. Before the onset of motor symptoms, there will be a period of non motor symptoms, including sleep disorders such as rapid eye movement (REM) sleep behavior disorder, which is an important predictive indicator for the progression of Parkinson's disease in the future. Therefore, studying individuals with rapid eye movement sleep behavior disorder provides an opportunity to gain a deeper understanding of the early pathological changes before the onset of Parkinson's disease. Co corresponding author of the paper, Jenny H. from the Ormond Street Institute for Children's Health at University College London, UK? Llqvist and Michael Bartl from the University of G ö ttingen Medical Center in Germany, along with colleagues and collaborators, analyzed blood samples from 99 recently diagnosed patients with Parkinson's disease, 72 patients with rapid eye movement sleep behavior disorder but no Parkinson's related motor symptoms, and 36 healthy controls. They identified sustained dysregulation of 23 proteins involved in inflammation, coagulation cascade, and Wnt signaling pathway from the blood of Parkinson's disease patients. Among these proteins, 6 also showed dysregulation in patients with rapid eye movement sleep behavior disorder. Subsequently, the authors of the paper used a machine learning model to predict diagnostic results based on protein composition. Based on the expression of 8 proteins, the model was able to identify 100% of Parkinson's patients. They then tested whether machine learning models could predict whether a patient with rapid eye movement sleep behavior disorder would develop Parkinson's disease. The results show that the model can predict the onset of Parkinson's disease up to 7 years before the onset of motor symptoms, with an accuracy of 79%. The authors of the paper summarize that identifying early Parkinson's disease patients can enable more people to participate in preventive clinical trials, improve patient treatment plans and research outcomes. Further research in this area needs to be validated in larger cohorts in order to translate these findings into clinical applications. (Lai Xin She)