Sas system for windows release 9

Data were analyzed using the MIXED procedure of SAS System for Windows Release 9.3 (SAS Institute Inc., Cary, NC). Data were analyzed as a completely randomized design. For insulin and time titration experiments for protein expression data, the model contained the effects of insulin and time, along with their interaction; to account for nonconstant variance in the data, all values were log transformed before being modeled. Each representative was modeled as a random blocking factor with a Kenwood‐Rodgers degrees of freedom adjustment. For insulin × rapamycin experiments for protein expression data and AAT mRNA abundance data, the model contained the effects of insulin and rapamycin, and along with an interaction term. Each representative was modeled as a random blocking factor. For insulin and time titration experiments for AAT mRNA abundance data, the model contained the effects of insulin and time, and along with an interaction term. Each representative was modeled as a random blocking factor with a Kenwood‐Rodgers degrees of freedom adjustment. Treatment means were computed using the LSMEANS option, and pairwise t‐tests were used to separate means and significance was determined at P < 0.05. All data are presented at the original scale with standard errors.