Machine Learning

Based on the rate at which the condition advances, Weill Cornell Medicine researchers have classified Parkinson's disease into three subgroups using machine learning. Apart from their potential utility as diagnostic and prognostic tools, these subtypes are identified by unique driver genes. If verified, these indicators may potentially point to potential pharmacological targets for the subtypes. Their findings were published in the journal Npj Digital Medicine. Parkinson’s disease is highly heterogeneous, which means that people with the same disease can have very different symptoms, his indicates there is not likely to be a one-size-fits-all approach to treating it. We may need to consider customized treatment strategies based on a patient’s disease subtype. Dr. Fei Wang The different patterns of illness development that the subtypes exhibited allowed the investigators to identify them. They designated them as the Moderate Pace subtype (PD-M, about 51% of patients), the Inching Pace subtype (PD-I, about 36% of patients) for disease with mild baseline severity and mild progression speed, and the Rapid Pace subtype (PD-R), for cases with the fastest rate of symptom progression. By utilising deep learning-based methods to examine de-identified healthcare information from two sizable datasets, they were able to determine the subtypes. Through the use of network-based techniques to analyse patient genetic and transcriptome profiles, they also investigated the molecular mechanisms connected to each subtype. For instance, some pathways, including those about neuroinflammation, oxidative stress, and metabolism, were activated in the PD-R subtype. Additionally, the three subtypes were discovered to have different brain imaging and cerebrospinal fluid indicators by the study. Since 2016, when the team took part in the Michael J. Fox Foundation-sponsored Parkinson's Progression Markers Initiative (PPMI) data challenge, Dr. Wang's lab has been researching Parkinson's disease. After winning the challenge of generating subtypes, the team has been supported by the foundation to carry out this research. In their study, they used the Parkinson's Disease Biomarkers Programme (PDBP) cohort of the National Institute of Neurological Disorders and Stroke (NINDS) to validate the data gathered from the PPMI cohort, which served as the main subtype development cohort. Utilizing these discoveries, the researchers were able to pinpoint potential medication options that may be modified to specifically target the molecular alterations observed in the various subtypes. They then verified that these medications may slow the course of Parkinson's disease using two sizable, real-world databases of patient medical records. The OneFlorida+ Clinical Research Consortium and the New York-based INSIGHT Clinical Research Network are two databases that are a component of the National Patient-Centered Clinical Research Network (PCORnet). Dr. Rainu Kaushal, chair of the Department of Population Health Sciences at Weill Cornell Medicine and NewYork-Presbyterian/Weill Cornell Medical Centre, and senior associate dean for clinical research at Weill Cornell Medicine, is the leader of INSIGHT. By examining these databases, we found that people taking the diabetes drug metformin appeared to have improved disease symptoms especially symptoms related to cognition and falls compared with those who did not take metformin. Dr. Chang Su This was particularly true for those with the PD-R subtype since they are more prone to experience cognitive impairments at an early stage of the illness. We hope our research will lead other investigators to think about using diverse data sources when conducting studies like ours, We also think that translational bioinformatics investigators will be able to further validate our findings, both computationally and experimentally. Dr. Fei Wang Also, Read| First identified in humans, CoRSIVs are now found in cattle Scientists from the Cleveland Clinic, Temple University, University of Florida, University of California at Irvine, University of Texas at Arlington, and doctorate candidates from Cornell Tech's computer science programme and Cornell University's Ithaca campus' computational biology programme were among the many collaborators on this work. Source: Weill Cornell Medicine News. Journal Reference: Su, Chang, et al. "Identification of Parkinson’s Disease PACE Subtypes and Repurposing Treatments through Integrative Analyses of Multimodal Data." Npj Digital Medicine, vol. 7, no. 1, 2024, pp. 1-22, DOI: https://doi.org/10.1038/s41746-024-01175-9. Accessed 16 Jul. 2024. Last Modified:

Category: Machine Learning

New subtypes of Parkinson’s Disease can be defined by Machine Learning

Areas Covered

Discover More

Roche AXELIOS 1: A Transformative Step for Next-Generation Sequencing Platforms Launched

AstraZeneca Datroway EU Approval: CHMP Recommends Novel ADC for Metastatic Triple-Negative Breast Cancer

Merck KGaA Acquires Bio-Techne to Advance ProteinSimple and RNAscope Workflows in $11.3 Billion Deal

Pfizer Inc. Announces Milestone as FDA Approves Pfizer’s Ibrance Regimen for Advanced Breast Cancer Frontline Maintenance

Eli Lilly Completes Acquisition of Centessa Pharmaceuticals to Advance Orexin Receptor 2 Agonist Therapies

Gilead Trodelvy European Commission Approval Marks Historic First-Line Breast Cancer Advance