MIC‐MAC: An automated pipeline for high‐throughput characterization and classification of three‐dimensional microglia morphologies in mouse and human postmortem brain samples

Abstract

The phenotypic changes of microglia in brain diseases are particularly diverse and their role in disease progression, beneficial, or detrimental, is still elusive. High‐throughput molecular approaches such as single‐cell RNA‐sequencing can now resolve the high heterogeneity in microglia population for a specific physiological condition, however, the relation between the different microglial signatures and their surrounding brain microenvironment is barely understood. Thus, better tools to characterize the phenotypic variations of microglia in situ are needed, particularly for human brain postmortem samples analysis. To address this challenge, we developed MIC‐MAC, a Microglia and Immune Cells Morphologies Analyser and Classifier pipeline that semiautomatically segments, extracts, and classifies all microglia and immune cells labeled in large three‐dimensional (3D) confocal image stacks of mouse and human brain samples. Our imaging‐based approach enables automatic 3D‐morphology characterization and classification of thousands of individual microglia in situ and revealed species‐ and disease‐specific morphological phenotypes in mouse aging, human Alzheimer's disease, and dementia with Lewy Bodie's samples. MIC‐MAC is a precision diagnostic tool that allows a rapid, unbiased, and large‐scale analysis of microglia morphological states in mouse models and patient brain samples.

MORPHIOUS: A Machine Learning Workflow to Naively Detect the Activation of Microglia and Astrocytes

“…Our approach has significant advantages over previous work that identify activated glia through clustering in feature space alone, such as through k-means or hierarchical clustering [3][4][5][6][7] . While previous methods can evaluate the putative activation of individual cells, the heterogeneity in microglia morphology poses a risk for false positives, which are difficult to interpret given the nature of unlabelled data.…”

“…To better understand glial activation, several machine learning approaches have been developed to classify activated states based on their morphology. Commonly, these methods deploy unsupervised learning algorithms (e.g., K-means clustering, hierarchical clustering) [3][4][5][6][7] . In general, these approaches aim to group activated and non-activated glial morphologies into distinct groups based on the similarities of their features.…”

“…In general, these approaches aim to group activated and non-activated glial morphologies into distinct groups based on the similarities of their features. However, given that the activation of microglia and astrocytes exists along a gradient of morphologies 4,[7][8][9] , it remains difficult to define a strict classification boundary to accurately identify activated cells.…”

“…MORPHIOUS collects features through the widely used software ImageJ. The use of features with greater complexity, as recently described by other methods4 , may further improve the range of classifications that can be identified. MORPHIOUS is adaptable to other unsupervised classifiers and could be further improved with state-of-the-art convolutional neural networks that can effectively interpret features from raw image files.In conclusion, we demonstrate that MORPHIOUS can naïvely identify clusters of morphologicallyactivated microglia and astrocytes.…”

Tissue and Stem Cell Sourced Extracellular Vesicle Communications with Microglia

“…Here "M1-like" or "M2-like" is used in accordance with the cited study findings as a reference to which portion of this spectrum the microglia are predominantly embodying. M1-like microglia, expressing CD86 + , CD206-, and CD16/32 + [20][21][22], produce cytokines such as INF-y, IL-6, TNF-a, IL1-B, KC/GRO/CINC. During activation, microglia also upregulate expression.…”