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.
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. Convolutional neural networks (CNN) were constructed using the PyTorch open-source deep learning library in the Python programming language (version 3.8.5). Unless otherwise indicated, all other analyses were performed in the R programming language and statistical computing environment (version 4.1.0).
Please see the analysis workflow below.
To read our documented code, please visit www.serranopozolab.org/glia-ihc.
Our full codebase is available for download on GitHub.