An effective spectral unmixing algorithm for flow cytometry based on GA and least squares

Area of Science:

• Biomedical Engineering
• Analytical Chemistry
• Computational Biology

Background:

• Spectral flow cytometry is crucial in biology, chemistry, and medicine.
• Accurate spectral unmixing is essential for analyzing complex samples.
• Conventional methods often require pre-measured pure spectra, limiting application scope.

Purpose of the Study:

• To develop an automated spectral unmixing algorithm for spectral flow cytometry.
• To overcome the limitations of conventional methods by eliminating the need for pre-measured pure spectra.
• To enhance the analysis of complex samples, including those with unknown components or auto-fluorescence.

Main Methods:

• A novel spectral unmixing algorithm utilizing a genetic algorithm.
• The genetic algorithm optimizes basic spectra positions and peak sharpness for unknown components.
• Least squares method is employed for simultaneous estimation of component concentrations.

Main Results:

• The algorithm successfully separates overlapping spectra without pre-measured pure spectra.
• Simulations demonstrated good convergence rate, accuracy, and stability under various conditions.
• Experimental validation using cyanobacteria flow spectra confirmed the algorithm's feasibility and effectiveness.

Conclusions:

• The proposed spectral unmixing algorithm offers a wider application scope compared to conventional methods.
• It is effective for analyzing multi-stained samples with unknown components and auto-fluorescence.
• This advancement improves spectral flow cytometry analysis in biological, chemical, and medical fields.