Machine Learning and Fluorescence Microscopy Uncover Cellular Light Stress Responses

Key Details

• 1Custom automated fluorescence microscopy tracked hundreds of individual algal cells under light stress.
• 2Machine learning (dictionary learning, LDA) was used to distinguish three distinct NPQ (non-photochemical quenching) components at the single-cell level.
• 3Strong single-cell-level trade-offs were observed between protective mechanisms qE and qT, which bulk measurements cannot detect.
• 4Approach was non-destructive and adaptable to other photosynthetic organisms and stress types with detectable fluorescence signals.
• 5Integration with single-cell omics, flow cytometry, and microfluidics is suggested for broader biological and biotechnological application.