Cyclic Multiplex Fluorescent Immunohistochemistry and Machine Learning Reveal Distinct States of Astrocytes and Microglia in Normal Aging and Alzheimer’s Disease

We have developed a methodology of cyclic multiplex fluorescent immunohistochemistry on human postmortem brain sections followed by an image analysis and machine learning pipeline that enables a deep morphological characterization of astrocytes and microglia in the Alzheimer’s brain.

Dependencies

To run our code, please install the following dependencies:

Additional required libraries are specified in each script. Image segmentation was performed with the FIJI distribution of the open-source Java-based image analysis program ImageJ.^1,2 Convolutional neural networks (CNN) were constructed using the PyTorch open-source deep learning library in the Python programming language (version 3.8.5).³ Unless otherwise indicated, all other analyses were performed in the R programming language and statistical computing environment (version 4.1.0).

Workflow

Please see the analysis workflow below. Click on any icon to be navigated to the appropriate page.

Documentation

To read our documented code, please visit www.serranopozolab.org/glia-ihc.

Code Availability

Our full codebase is available for download on GitHub.