The methylated and non-methylated locus after bisulfite modification differs in number of cytosines (for detail see: What is bisulfite modification?).

Different base composition of the template may lead to the differences in the efficiency of PCR amplification of those two versions of a locus and this phenomenon is referred to as PCR bias. PCR bias in methylation studies was first identified by Warnecke PM et al. PMID:9336479.

This phenomenon was shown to present a serious pitfall in PCR based methylation detection leading to underestimation of methylation content of the screened sample.

PCR bias presents a serious challenge especially in experiments that aim to detect methylated alleles in a sample where it is present at low frequency in a background of non-methylated alleles.

In those samples higher efficiency of PCR amplification of the non-methylated allele may even lead to lack of ability to detect the methylated allele and consequently false negative result, described in: PMID:19483476.

Sample material with low content of methylated alleles of interest and a high background of non-methylated allele are for example:

• whole blood where only a specific type of cells may contain the methylated allele of interest and the vast majority of the cells contain non-methylated alleles.
• FFPE samples where the methylated allele in the DNA sample is underrepresented due to significant DNA degradation.
• liquid biopsies where only a few copies of the methylated allele may be present due to the nature of the sample.

The innovative primer design for Methylation-Sensitive High-Resolution Meting technology (MS-HRM) allows to overcome PCR bias and enables highly sensitive methylation detection even in samples where the methylated allele is significantly underrepresented.

For details of primer design see: PMID:18710507