MDS Abstracts

Abstracts from the International Congress of Parkinson’s and Movement Disorders.

MENU 
  • Home
  • Meetings Archive
    • 2024 International Congress
    • 2023 International Congress
    • 2022 International Congress
    • MDS Virtual Congress 2021
    • MDS Virtual Congress 2020
    • 2019 International Congress
    • 2018 International Congress
    • 2017 International Congress
    • 2016 International Congress
  • Keyword Index
  • Resources
  • Advanced Search

Variable selection using machine-learning to identify new signatures of patient-derived aggregated α-synuclein-induced neurodegeneration in non-human primates

M. Bourdenx, A. Nioche, S. Dovero, M.L. Arotçarena, S. Camus, G. Porras, M.L. Thiolat, N. Rougier, A. Prigent, P. Aubert, S. Bohic, N. Kruse, B. Mollenhauer, S. Novello, M. Morari, I. Trigo, M. Goillandeau, M. Tasselli, C. Perier, N. Garcia Carrillo, C. Estrada, A. Recasens, J. Blesa, M. Herrero, P. Derkinderen, M. Vila, J. Obeso, B. Dehay, E. Bezard (Bordeaux, France)

Meeting: 2018 International Congress

Abstract Number: 1683

Keywords: Alpha-synuclein, Prion diseases. See Transmissible spongiform encephalopathies

Session Information

Date: Monday, October 8, 2018

Session Title: Parkinson's Disease: Pathophysiology

Session Time: 1:15pm-2:45pm

Location: Hall 3FG

Objective: Emerging evidence strongly suggests that α-synuclein, a major protein component of LB, may be responsible for the spreading of the pathological process within affected individuals. Recently, through an innovative strategy based on the purification of Lewy bodies (LB) containing aggregated α-synuclein from the substantia nigra pars compacta of PD patients, we assessed the prion-like properties of endogenous α-synuclein assemblies in wild-type mice and non-human primates (Recasens et al., Ann. Neurol. 2014).

Background: The pilot nature of the demonstration however called for a properly powered demonstration in non-human primates, which was the aim of this study, achieving in a large group of baboons (n=49).

Methods: After in vitro and in vivo (in wild-type mice) LB-induced toxicity validation, α-synuclein-containing extracts were injected bilaterally into the striatum (either a mixture of LB fractions or no-LB fractions derived from the same 3 PD patients, which contains soluble or finely granular a-synuclein but lacks large LB-linked a-synuclein aggregates.

Results: After a live phase of 2 years, extensive analysis was performed using biochemical and histochemical techniques in the whole brain. This study collected over 180 variables in each monkey. To overcome the roadblock associated to the “p > n” problem that occurs when the number of variables measured is greater than the sample size, we developed a multiple layer perceptron (MLP), i.e. an artificial neural network commonly used in machine learning. The performance of a given combination of variables was measured to predict the level of degeneration and extract meaningful variables. Variables were then sorted according to their occurrence in the top 1% of the best combinations.

Conclusions: This MLP allowed to identify two types of variables: the ones that reflect the actual neurodegeneration – the variables that describe the phenomenon to be explained – and the ones that might contribute to the pathogenic mechanism – the variables that could be useful to explain the phenomenon. Overall, this study using this unbiased methodology, confirmed highly-expected variables but, more importantly, also identified unexpected variables that appear to be excellent predictors for dopaminergic neurodegeneration.

References: A. Recasens, B. Dehay, J. Bové, I. Carballo-Carbajal, S. Dovero, A. Pérez, P.O. Fernagut, J. Blesa, A. Parent, C. Perier, I. Fariñas, J.A. Obeso, E. Bezard and M. Vila.. Lewy Body extracts from Parkinson’s Disease Brains trigger α-Synuclein Pathology and Neurodegeneration in Mice and Monkeys Annals of Neurology 2014, 75:351-62.

To cite this abstract in AMA style:

M. Bourdenx, A. Nioche, S. Dovero, M.L. Arotçarena, S. Camus, G. Porras, M.L. Thiolat, N. Rougier, A. Prigent, P. Aubert, S. Bohic, N. Kruse, B. Mollenhauer, S. Novello, M. Morari, I. Trigo, M. Goillandeau, M. Tasselli, C. Perier, N. Garcia Carrillo, C. Estrada, A. Recasens, J. Blesa, M. Herrero, P. Derkinderen, M. Vila, J. Obeso, B. Dehay, E. Bezard. Variable selection using machine-learning to identify new signatures of patient-derived aggregated α-synuclein-induced neurodegeneration in non-human primates [abstract]. Mov Disord. 2018; 33 (suppl 2). https://www.mdsabstracts.org/abstract/variable-selection-using-machine-learning-to-identify-new-signatures-of-patient-derived-aggregated-%ce%b1-synuclein-induced-neurodegeneration-in-non-human-primates/. Accessed June 14, 2025.
  • Tweet
  • Click to email a link to a friend (Opens in new window) Email
  • Click to print (Opens in new window) Print

« Back to 2018 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/variable-selection-using-machine-learning-to-identify-new-signatures-of-patient-derived-aggregated-%ce%b1-synuclein-induced-neurodegeneration-in-non-human-primates/

Most Viewed Abstracts

  • This Week
  • This Month
  • All Time
  • Humor processing is affected by Parkinson’s disease and levodopa
      • Help & Support
      • About Us
      • Cookies & Privacy
      • Wiley Job Network
      • Terms & Conditions
      • Advertisers & Agents
      Copyright © 2025 International Parkinson and Movement Disorder Society. All Rights Reserved.
      Wiley