Cognitive impairment is a debilitating symptom in Parkinson’s disease (PD). We aimed to establish an accurate multivariate machine learning (ML) model to predict cognitive outcome in newly diagnosed PD cases from the Parkinson’s Progression Markers Initiative (PPMI). Annual cognitive assessments over an 8-year time span were used to define two cognitive outcomes of (i) cognitive impairment, and (ii) dementia conversion. Selected baseline variables were organized into three subsets of clinical, biofluid and genetic/epigenetic measures and tested using four different ML algorithms. Irrespective of the ML algorithm used, the models consisting of the clinical variables performed best and showed better prediction of cognitive impairment outcome over dementia conversion. We observed a marginal improvement in the prediction performance when clinical, biofluid, and epigenetic/genetic variables were all included in one model. Several cerebrospinal fluid measures and an epigenetic marker showed high predictive weighting in multiple models when included alongside clinical variables.

Welcome to the github repository containing all the scripts and several documents used in the process of developing and finalizing the work seen in the paper. If you're interested in the used scripts to process the methylation data and QC, machine learning, handling of the phenotype or any of our figure generation scripts, check out the corresponding links. If there are any questions, don't hesitate to ask!

Maastricht University Data Science Research Seminar Series (11-2022) - info - >Download slides<

University of Exeter Medical School ARE event (07-2022) - >Download slides<

Ehsan Pishva (ehsanpishva) - [email protected]

Joshua Harvey (JoshHarveyGit) - Undertook data analysis, support with data review, wrote the first draft of the manuscript.

Rick A Reijnders (Rrtk2) - Undertook data analysis, support with data review, wrote the first draft of the manuscript.

Ehsan Pishva (ehsanpishva) - Conceived and directed the project, wrote the first draft of the manuscript.

Annelien Duits, Byron Creese - Were involved in the selection of the clinical predictors and outcome. Rachel Cavil, Sebastian Köhler, Ali Torkamani - Provided advice on data analysis. Gemma Shireby - Contributed to generating polygenic scores.

Joshua Harvey, Rick A Reijnders, Ehsan Pishva, Katie Lunnon, Albert FG Leentjens, Lars Eijssen, Bart PF Rutten, Byron Creese, Annelien Duits - Contributed to the interpretation of the results.

All authors provided critical feedback on the manuscript and approved the final submission.

Finalized, accepted at npj parkinson's disease.

At ResearchSquare or at medRxiv

All code and documents in the PPMI-ML-Cognition-PD folder was created by these author(s).

The project's source code is freely reusable under the MIT License.