Objective Based on the data detected using gene expression microarray, to select multiple reference genes and use them to quantify transcriptome genes of different abundance in the mouse liver tissue. Methods To detect global transcriptome genes in the mouse liver tissues using gene expression microarray. All the genes were sorted into different groups according to their expression level, followed by coefficient of variation (CV) analysis. Real-time quantitative polymerase chain reaction (qPCR) and geNorm software were used to further identify the reference genes. Results The expression levels of over 60,000 genes were obtained from microarray screening, and divided them into low, moderate and high expression groups. Finally six reference genes were screened, i.e. Casp2 and Lrrc14 as low abundance, Nrd1 and Trpc4ap as moderate abundance, and Atp5a1 and Clu as high abundance genes. Conclusions The six reference genes derived from microarray data can be used to accurately quantify the global transcriptome genes with various expression levels.