Generalizable, high-throughput image analysis of subcellular structures using dispersion indices.

Area of Science:

• Cell Biology
• Bioimaging
• Computational Biology

Background:

• Subcellular structural dynamics are crucial for cell function.
• Current methods for quantifying these changes are labor-intensive and difficult to scale.
• A need exists for efficient and generalizable tools to analyze subcellular organization.

Purpose of the Study:

• To introduce GRID (generalized readout of image dispersion), an open-source algorithm for quantifying subcellular structural dynamics.
• To demonstrate the algorithm's generalizability across various cellular structures and model systems.
• To enable high-throughput analysis for disease research and drug discovery.

Main Methods:

• Developed GRID, an algorithm applying dispersion indices to quantify diffusiveness and aggregation of subcellular structures from image data.
• Applied GRID to analyze autophagic puncta, mitochondrial clustering, and microtubule dynamics.
• Validated GRID in 2D and 3D cell cultures, including multicellular human midbrain organoids.

Main Results:

• GRID efficiently quantifies subcellular structural dynamics and spatial locations.
• Demonstrated successful detection of changes in autophagy, mitochondria, and microtubules.
• Identified cell-type specific autophagy responses in human organoids.
• Enabled estimation of half-maximal effective concentration (EC50) values.

Conclusions:

• GRID offers a computationally efficient and generalizable approach for high-throughput subcellular analysis.
• The algorithm facilitates understanding of disease mechanisms in areas like metabolic disorders, neurodegeneration, and cancer.
• GRID serves as a valuable tool for initial screening in drug discovery efforts.