“…For example, colon and prostate tumor tissues often contain uneven proportions of muscle cells, which are known to be rich in circRNAs [ 99 ]. Consequently, the discrepancy in muscle cell content between different samples may give rise to an artifactual bias in circRNA expression [ 85 , 86 , 98 , 99 ]. Similarly, lncRNA expression is also cell-specific, which means that the analysis of lncRNA/circRNA expression and function using large tissue samples, such as tumor and non-tumor adjacent specimens, may result in erroneous conclusions.…”