Fig. 2

a,b, HyU (a) and LU (b) tested using a hyperspectral fluorescence simulation that was generated from four fluorescent signatures (emission spectra; Supplementary Fig. 5e). c, Absolute MSE shows that HyU offers a consistent reduction in error across a broad range of photons per spectra (no. photons per independent spectral components, here resulting from four reference spectra combined). Shaded regions for line plots denote the 95% confidence interval around the mean. d, The performance differences in the MSE of HyU relative to LU persists when applying multiple phasor denoising filters (0–5 median filters). Shaded regions for line plots denote the 95% confidence interval around the mean. e, Unmixing of experimental data from a four-color zebrafish shows increased contrast for HyU (left) compared to LU (right). Scale bar, 50 µm. f,g, The increased accuracy is revealed by residual images of HyU and LU, showing the spatial distribution of unassigned signals after the analysis of data in e. h, Box plots of the residuals (shown in f,g) presents values of 11% for HyU compared to 77% for LU with *(P < 10⁻¹⁰) with an independent two-sample t-test and n = 1.05 × 10⁶ pixels. Center shows median; box represents first and third quartiles; whiskers represent 1.5× first and third quartiles; min/max are not shown. i–p, Enlarged rendering of HyU results (e, white box) clearly shows low levels of bleed-through between labels (m–p). Similar enlargement of LU results show noticeably worse performance. Note that regions with bright signals (membrane j,n, white arrow) bleed through other channels (m,o). Scale bar, 20 µm. Tetra-labeled specimen used here was Gt(cltca-citrine);Tg(ubiq:lyn-tdTomato; ubiq:Lifeact-mRuby;fli1:mKO2). The sample depicted in e is representative of 28 experimental sessions each with three to five biological replicates, yielding similar results.