New Host Specific Local Clock Model Predicting Future Generations

by Ralph Navarro, Lawrence Zeldin, and Soo Yeon (Nancy) Kim

The Influenza A virus (IAV) has been a serious burden on public health, possibly since before recorded history. While modern medicine has led to vaccines to many other viruses, the high mutational rate of Influenza A renders any created vaccine obsolete within a short period of time. This mutation, and the virus’s ability to jump between hosts of various species, also leads to sporadic strains that are highly virulent and cause pandemics. In order to predict these pandemics, several models have been created by various researchers that use an assumed mutational rate to predict future outcomes, however, these models were met with limited success. In “A synchronized global sweep of the internal genes of modern avian influenza virus”, researchers showed that assuming that mutational rates depended on the host type allowed for the creation of a more effective model, and used this model to help explain several patterns seen in IAV evolution.

There have been difficulties in making the phylogenetic analyses of viral movement timing and direction to be accurate because IAV in each host species has a different evolutionary rate. Past models assumed a single mutational rate across all IAV strains, and were unsuccessful at effectively modelling viral evolution. In order to solve this problem, the scientists created the host-specific local clock (HSLC) model which could keep the accuracy of phylogenetic analyses by taking varying mutational rates across hosts into account. The different host groups, ‘equine,’ ‘human’ and ‘avian’ were all labeled groups 1, 2 and 3 with different mutational rates. The first trial for different models all showed similar results compared to the multiple trials tested after. The results of the first trial indicates that host-specific local clock (HSLC) could be used to predict the future generations of the three strains, ‘equine,’ ‘human’ and ‘avian.’

This model showed good consistency between predicted and observed patterns. For instance, the analyses of the full-length segments that code for the IAV polymerase proteins and thesurface glycoprotein subtypes proved how the HSLC model can perform better then the other models with a set specific mutational rate. The model also matched the predicted Uracil content of equine strains, which is positively correlated with how long a strain has been established in a mammalian host.

Using this model, the researchers analyzed the evolutionary origin of several IAV strains, including that of the 1918 pandemic. One observations was that in comparison to other genomic segments, the external genes HA and NA showed significantly higher diversity in wild type birds. These results suggested that an avian virus, such as H7N7 from domestic birds, started a global sweep of its internal genes, limiting the diversity of all the previous IAV internal genes by outcompeting them. Diversity of the HA and NA genes, however, would have been conserved, since antigen diversity would offer an advantage against host immune responses.

Using their more accurate model, the researchers were able to better predict the evolutionary origins of modern influenza strains, and were able to predict the how this evolution was shaped by a sweep of internal gene segments. And the ability to match the past evolution of the virus with this effective model may also shed some light upon future evolution of the virus, possibly allowing better predictions of future influenza outbreaks.

Worobey M, Han G, and Rambaut A. A synchronized global sweep of the internal genes of modern avian influenza virus. Nature508, 254–257. 

Link to article


Leave a Reply

Your email address will not be published. Required fields are marked *