Skip to content
← Back to browse
BRAIN Publication-derived

Machine learning-assisted neurotoxicity prediction in human midbrain organoids

Source Monzel et al., 2020 · Luxembourg Centre for Systems Biomedicine (LCSB), University of Luxembourg · 10.1016/j.parkreldis.2020.05.011

👤 Anna S. Monzel, Kathrin Hemmer, Tony Kaoma, Lisa M. Smits, Silvia Bolognin, Philippe Lucarelli, Isabel Rosety, Alise Zagare, Paul Antony, Sarah L. Nickels, Rejko Krueger, Francisco Azuaje, Jens C. Schwamborn ⏱ 48 days 📋 11 phases 🧫 Human iPSC

Abstract

This protocol describes the generation and neurotoxin treatment of human midbrain organoids derived from hiPSCs, combined with high-content imaging and machine learning-based analysis to predict neurotoxic effects on dopaminergic neurons. The method enables quantification of dopaminergic neuron death and neuronal complexity following 6-hydroxydopamine (6-OHDA) exposure, providing a platform for toxicity screening and Parkinson's disease modeling.

Cell source
Human iPSC
Application
Disease modeling; Neurotoxicity screening

Protocol overview

28 steps across 11 phases

Organoid generation and culture Days 0-35
  1. 1 Generate human midbrain organoids from hiPSCs
  2. 2 Culture organoids for five weeks prior to treatment
6-OHDA dose-response optimization Days 35-41 (6 days post-treatment)
  1. 1 Prepare 6-OHDA solutions at varying concentrations
  2. 2 Expose organoids to 6-OHDA for 48 hours
  3. 3 Return organoids to normal culture conditions
Organoid sample preparation for analysis Days 41-42
  1. 1 Fix organoids for immunofluorescence staining
  2. 2 Prepare organoid samples for protein extraction
  3. 3 Dissociate organoids into single cells
Flow cytometry analysis Days 41-42
  1. 1 Perform flow cytometry to quantify tyrosine hydroxylase-positive cells
  2. 2 Construct dose-response curves
Immunofluorescence staining and imaging Days 41-43
  1. 1 Section fixed organoids
  2. 2 Perform multiplex immunofluorescence staining
  3. 3 Acquire high-content imaging data
Western blot analysis Days 41-43
  1. 1 Extract protein from snap-frozen organoids
  2. 2 Perform Western blot for tyrosine hydroxylase
Image processing and feature extraction Days 43-45
  1. 1 Process high-content imaging data in MATLAB
  2. 2 Extract neuronal complexity features
  3. 3 Quantify TUJ1+ and TH+ cell populations
Statistical analysis and variance decomposition Days 45-46
  1. 1 Assess contribution of experimental factors using Principal Variance Component Analysis (PVCA)
  2. 2 Perform z-score normalization
Machine learning model development and validation Days 46-47
  1. 1 Build random forest classifier on raw (unprocessed) data
  2. 2 Build optimized random forest classifier on normalized data
  3. 3 Validate model performance metrics
Hierarchical clustering and principal component analysis Days 47-48
  1. 1 Perform hierarchical clustering on normalized data
  2. 2 Perform principal component analysis on scaled normalized data
Validation on PD patient-derived organoids (optional extension) Separate experiment
  1. 1 Obtain PD patient-derived midbrain organoids with LRRK2-G2019S mutation
  2. 2 Perform multiplex immunofluorescence for FOXA2 and TH
  3. 3 Extract features and build time-point-specific ML classifiers

Full SOP

🔬

Create a free account to access this protocol

Join OrganMatch to unlock step-by-step procedures, reagent concentrations, QC checklists, and downloadable batch record templates.

Create free account

Already registered? Log in

Attribution

This SOP was authored by Organthis based on the published method in Monzel et al., 2020. The originating laboratory holds no rights in this SOP and has not endorsed it unless marked Verified.

This wording is awaiting legal review.

Something wrong with this entry? Report an issue with this protocol

Need a commercial licence?
Use this protocol in your therapeutic or diagnostic pipeline.
Request Commercial Licence