Subcellular Feature-Based Classification of α and β Cells Using Soft X-ray Tomography

Area of Science:

• Cell Biology
• Biophysics
• Diabetes Research

Background:

• Pancreatic islet alpha and beta cell dysfunction is linked to diabetes.
• Subcellular organization in islet cells during disease progression remains poorly understood.
• Current 3D mapping methods are time-consuming and subjective.

Purpose of the Study:

• To develop an advanced method for identifying and characterizing islet cells.
• To investigate subcellular differences between alpha and beta cells.
• To explore cellular heterogeneity in pancreatic islets.

Main Methods:

• Developed a subcellular feature-based classification approach.
• Utilized soft X-ray tomography (SXT) for high-resolution imaging.
• Applied supervised machine learning and Uniform Manifold Approximation and Projection (UMAP) for analysis.

Main Results:

• Identified significant morphological and organelle differences between alpha and beta cells.
• Characterized biophysical differences in insulin and glucagon vesicles (size, molecular density).
• Successfully predicted cell types using sub-vesicular parameters and visualized cell subtypes.

Conclusions:

• The new SXT-based classification method accurately distinguishes islet cell types.
• Sub-vesicular biophysical properties offer novel insights into cellular heterogeneity.
• This methodology provides a powerful tool for studying cellular systems and disease progression.