Computational Simulation of Progranulin for
Blood-Based Detection of Parkinson’s Disease up to 7 Years Before Symptom Onset

Abstract

Parkinson’s disease (PD) is a neurological degenerative disease that affects signal pathways in the brain, leading to motor control issues, as well as various other conditions. Detecting PD is challenging because its symptoms develop slowly and are initially subtle. PD’s symptoms overlap with different neurological disorders, allowing it to be easily confused with disorders like Essential Tremor or Multiple System Atrophy (MSA). However, there may be a way to detect it years before it worsens. Recently, the granulin protein has been linked to the onset of PD, suggesting it could serve as a biomarker. Our research aims to identify granulin precursor proteins that are cleaved into granulin proteins. “Antibody” is a general term used to classify proteins that bind to various structures, allowing the body to identify and/or destroy them. We hypothesize that antibodies binding to the protein’s surface could be used to detect and diagnose PD. We will use HDOCK to find binding sites between the granulin precursor and a specific antibody. We will use Prodigy for binding energy. We will use P2Rank to predict the best binding sites on the granulin precursor. Based on the visual inspection and binding energy, we have selected antibody 1IGT as the most appropriate candidate for granulin protein detection. By finding the antibody that best binds to the granulin precursor, we can modify it to help us locate it. This would allow doctors to diagnose PD much earlier and provide faster, more effective treatment.