BACKGROUND: Although relative quantification by real-time PCR may be easier to perform than absolute quantification, there is a risk of errors associated with standard curve construction. The aim of this study was to evaluate the effects of standard curves on relative quantification using real-time PCR in Candida albicans. METHODS: The reproducibility of real-time PCR-based standard curves for target genes and a reference gene generated from PCR amplicons (10-fold serial dilution, 10(-4) to 10(-9)) was evaluated. In addition, the effects on standard curves were evaluated by running the same cDNA samples. RESULTS: The within-run variation (CV) by crossing point (Cp) was 0.12-1.05% for ERG11 and ACT1, whereas the between-run CV was 2.07-6.84% for ERG11, CDR1, MDR1 (target gene) and ACT1 (ref-erencegene). The differences in PCR efficiency between targets and reference may be attributable to variations in relative quantification. CONCLUSIONS: To achieve reliable relative quantification of mRNA in real-time PCR, a feasible guide-line and standardization are of major importance.