The paper makes the case for a reassortment pattern of HA, PB1 and PB2 that is very different from that of the other genome segments. I am not sure if this conclusion is fully convincing. While there is certainly a large difference in the TMRCA between segments (fig. 3), this difference may be based on very few large reassortment events. For example, the segment NA also shows very deep branches (Fig. 2), with the replacement of Yam by Vic on the lower branch apparently caused by a single successful reassortment in 2003, followed by an extinction of the donor branch. This segment could be argued to be an intermediate case between HA, PB1 and PB2 and the other segments.
The paper makes the case for a reassortment pattern of HA, PB1 and PB2 that is very different from that of the other genome segments. I am not sure if this conclusion is fully convincing. While there is certainly a large difference in the TMRCA between segments (fig. 3), this difference may be based on very few large reassortment events. For example, the segment NA also shows very deep branches (Fig. 2), with the replacement of Yam by Vic on the lower branch apparently caused by a single successful reassortment in 2003, followed by an extinction of the donor branch. This segment could be argued to be an intermediate case between HA, PB1 and PB2 and the other segments.