Extended Data Fig. 5: Generalized Linear Model results.

We fitted a generalized linear model (estimated using ML) to predict rhodopsin abundance and derived a reduced minimal model based on Akaike Information Criterion (AIC) values using a backwards stepwise algorithm. Both models’ explanatory power is substantial (R2 = 0.40). Standardized parameters were obtained by fitting the models on a standardized version of the dataset. 95% Confidence Intervals (CIs) and p-values were computed using a Wald t-distribution approximation.