Most cervical cancers are caused by persistent infections with one of a dozen carcinogenic human papillomaviruses (HPV). The cancer risk differs by HPV genotype, with HPV16 and HPV18 accounting for over 70% of all cervical cancers. E6 and E7, are potent oncoproteins that inhibit apoptosis by disabling p53 and activate the cell cycle by disrupting the pRB pathway. In addition, the viral oncogenes cause major chromosomal instability already at precancerous stages and may induce integration of the viral genome into the host cells. Despite these strong oncogenic features of HPV, only a small subset of HPV infections ever progress to precancer or cancer. Recently, the important role of host and viral methylation in the molecular etiology of cervical cancer has been recognized. We recently conducted a discovery effort that yielded a set of host gene methylation markers that showed promising performance to detect cervical precancer. Similarly, we recently showed that regions of the HPV genome of 11 carcinogenic types are highly methylated in women with CIN3 compared to women with transient infections, suggesting that methylation is a general phenomenon in the transition from infection to precancer. Methylation patterns were found to be similar between closely related HPV genotypes. These findings may improve our understanding of the molecular basis for viral persistence and increase our ability to identify women who are more likely to progress to cervical cancer. Measuring host or HPV DNA methylation could serve as a specific marker for cervical precancer in cervical cancer screening.