UASScer-driven RNAi transgenes (either TRIP or VDRC lines) were used to silence candidate UPR and ER stress-related genes in a fic30C/+heterozygous background, with either LongGMR-Gal4, UAS-FicE247G or LongGMR-Gal4 only. Eye roughness was scored for individual flies and averaged for each cross. Table reports number of flies scored in each group (0 = no roughness, 2 = mildly rough (control flies), 4 = severely rough, 1 and 3 are intermediate phenotypes) and the weighted average of the eye roughness. Significance differences are highlighted in yellow, and p-values were determined using Fisher’s Exact Test for categorical data, comparing the effects of each gene knockdown with the control group (top line, fic30C/+; LongGMR-Gal4, UAS-FicE247G). Interactions were considered significant for any individual test if p<0.003 as determined using Bonferroni’s multiple comparison adjustment.