An Ignored Pitfall in Selecting Reference Genes and the Solution with Emphasis on Subtypes and Sex of Thyroid Cancer Patients
Abstract Routine tissue specific reference genes are widely used in transcriptomics studies without concerning their target genes, sex of patients, and diseases subtype. We proposed the concept of specific reference genes for each target gene after considering sex of patients and thyroid cancer subtypes. RT-qPCR technique was coupled with expression meta-analysis of samples with different races and ethnicities, in both sexes, and in different thyroid cancer subtypes. Eight common reference genes were evaluated and some of them undoubtedly ruled out. We found that mean and SD values of the genes must be considered carefully before the selection of reference genes. A formula was also developed accordingly and we equipped it with statistical analysis of more than 25000 genes. In conclusion, the reckless selection of reference genes can distort the output and must be prohibited.