Abstract
ABSTRACT
Cells maintain homeostasis by dynamically reorganizing their organelles to tune metabolism in response to stress. Fluorescence microscopy maps organelle structures with subcellular resolution but provides limited chemical information. Infrared (IR) imaging offers a label‐free alternative for probing intrinsic molecular vibrations that report on lipids, carbohydrates, and nucleic acids. Here, we combine submicron optical photothermal IR imaging with machine learning to classify subcellular structures in fixed U‐2 OS cells in a cell state dependent manner. Using fluorescent‐labeled organelles as ground truth, we trained and evaluated random forest (RF) classifiers and U‐Net convolutional neural networks to identify organelles from IR spectra in healthy cells. The classifiers accurately identified multiple organelles, including the endoplasmic reticulum, Golgi apparatus, mitochondria, nucleus, nucleolus, and stress granules. Classifiers trained in U‐2 OS cells transferred without retraining to HEK 293 cells, consistent with conserved organelle biochemical composition across cell types. Conversely, the classifiers failed under cellular stress which indicates sensitivity to stress‐induced changes in organelle state. Together, these results establish a scalable, label‐free strategy for high‐resolution mapping of organelle biochemical composition and provide a foundation for subcellular biomarker discovery and disease‐state diagnostics.