Fig. 7

The workflow of this study. a The sample information of transcriptome sequence data of these IMCG (n = 3*3), DBG (n = 3*3) and F1_Adult (n = 7*3); b Candidate housekeeping genes were preliminarily selected by four indicators which including FPKM, CV, DPM, and MFC, and were further selected by Venn diagram analysis; c The sample information of the qRT-PCR experiments on the 4 experimental groups, with 3 biological replicates in every level of each group. Group 1, different development stages (n = 4*3); Group 2, hair follicle cycle stages (n = 3*3); Group 3, breeds (n = 3*3); Group 4, sampling sites (n = 5*3). d Candidate housekeeping genes were determined using 4 algorithms, including geNorm, NormFinder, BestKeeper and the ΔCt method. An additional comprehensive analysis was conducted using ComprFinder, a new algorithm developed by the authors. e The selected HKGs were validated by 3 target genes, and were performed on 18 skin samples (n = 6*3)