Proc mixed of sas version 9

ERG amplitudes were converted to natural logarithms to better approximate normal distributions. JMP Pro, version 11 (SAS Institute, Cary, NC) was used to compare cross-sectional ERG amplitudes and implicit times based on interocular difference scores, which were normally distributed by the Shapiro-Wilks W Goodness of Fit test. JMP Pro with Dunnett's method was used to compare RPGR expression divided by rootletin expression between wt mice (as control), mice injected with the short construct, and mice injected with the long construct; these ratio data were normally distributed. Repeated – measures regression with PROC MIXED OF SAS, version 9.3 (SAS Institute) or JMP Pro were used to compare ONL thicknesses and inner segment/outer segment lengths for treated versus untreated eyes at two ages. When these histologic data departed from normality, they were converted to normalized ranks by the Van der Waerden transformation and the regression analyses repeated. Repeated-measures regression was also used to compare rates of change of ERG amplitude over follow-up by eye. Using log data allowed these longitudinal ERG amplitudes to be fitted to a linear model consistent with the published exponential loss of ERG amplitude over time in mice with hereditary retinal degenerations^{46 (link)}.