Fig. 3

Identifying a spectral difference between groups. a Estimation of two main spectral components of within-subject variability (w1, green and w2, purple). These components were largely explained by a combination of spectra of known ocular constituents and were used to correct each reflectance spectrum. b Corrected reflectance spectra at sampling location S1 (n = 20 controls, n = 15 cases, data shown as mean ± SEM) revealed a spectral difference between the two groups. c P values for two-sided unpaired t tests between groups using false discovery rate (FDR) control for significance across all the wavelengths (n.s. is for non-significant). d Spectral model at sampling location S1 corresponding to the main spectral difference between the two groups