Per-pixel unmixing of spectrally overlapping fluorophores using intra-exposure excitation modulation: Software and datasets

Here we provide the sourcecode and data used in this work that was accepted for publication in the journal Talanta:

Per-pixel unmixing of spectrally overlapping fluorophores using intra-exposure excitation modulation

Hana Valenta^{1,*}, Franziska Bierbuesse1^{1,*}, Raffaele Vitale^2, Cyril Ruckebusch^2, Wim Vandenberg^1, Peter Dedecker^1

1 Department of Chemistry, KU Leuven, Belgium

2 U. Lille, CNRS, LASIRE, France

* These authors contributed equally to this work.

The sourcecode was written for MATLAB R2022a and is provided in form of the MATLAB script Unmixing_software.m and bases on non-negative least squares. The latter is given with the fnnls function provided by the MCR-ALS toolbox (http://www.mcrals.info). We are grateful to Joaquim Jaumot, Romà Tauler and Anna de Juan for providing the MCR-ALS toolbox.

All used datasets from the publication are provided. Single fluorophore datasets are given in the Single_fluorophore folder for ffDronpa targeted to the nucleus, EGFP to the plasma membrane, ffDronpa2F to peroxisomes and ffDronpaF to vimentin-based fibers. Datasets with a mixture of all four fluorophores are provided in the folder MIX_fluorophore. After downloading, the datasets need to be extracted before using the MATLAB analysis procedure. If you wish to use the analysis procedure with other datasets, the same folder structure needs to be implemented.

To run the procedure, open Unmixing_software.m with MATLAB and run it. Always ensure that at the beginning of the analysis the MATLAB current folder corresponds to the folder containing Unmixing_software.m, fnnls.m and the data folders. If MATLAB asks you to change the MATLAB current folder to the location of the MATLAB routine, accept to change it. The datasets will be read in automatically. Results will be saved in the Results folder.

Output:

Single_Fluorophore\Profiles\FluoProfile-(fluorophore)_norm_witherror:   Fluorescence profiles for a single "fluorophore"

MIX_fluorophore\Intensity_(dishname)_FOV(#) - (fluorophore):   Unmixed intensity images for "dishname", FOV number "#" and "fluorophore" for mixed datasets

Single_Fluorophore\(fluorophore)\ScatterConfusionMatrix:   Contribution plots for unmixed images for single fluorophore datasets

The contribution plots are only calculated if the user selects "yes" in the MATLAB menu during running the procedure. The runtime will increase with this part (in our case by about 15 minutes).