Applications of Next-Generation Sequencing Technologies to Diagnostic Virology

Barzon, Luisa; Lavezzo, Enrico; Militello, Valentina; Toppo, Stefano; Palù, Giorgio

Populations of DNA viruses are considered less complex and variable when compared to RNA viruses. However, data from deep sequencing of DNA virus genomes have revealed that complex mixtures of viral genotypes may be present in infected subjects and that positive selection could have contributed to the divergence of different strains. This is the case of human cytomegalovirus (HCMV), which establishes lifelong latent infections in humans and may reactivate and cause severe life-threatening disease in immunocompromised patients. High intra-host variability of HCMV genome was demonstrated in lung transplant recipients by deep sequencing of the amplicons of three variable HCMV genes [92] and in neonates with congenital HCMV infection by deep sequencing of long range, overlapping amplicons covering the entire HCMV genome [93] . Since PCR amplification and sequencing can introduce errors in their own, which could be misinterpreted as mutation or polymorphisms, deep sequencing studies have to develop protocols and algorithms to estimate experimental error and to filter false positive results. In the studies reported here on HCMV genome variability, experimental error rate was estimated by using arbitrary criteria [92] or an algorithm based on experimental data obtained from deep sequencing analysis of a control HCMV genome cloned in a BAC vector [93] .