Result
The H1 phylogeographic tree showed two large independent clusters within Iran, suggesting the initiation of two independent effective outbreaks with different tMRCAs, one in the past years (1997) and the other in recent years (2005). The first outbreak was initiated from the north region of the country and most of time was circulated between the north and northwest region within the country and established some small introduction in the southwest region in the late years (2008). The phylogeographic tree showed that the second initiation was introduced in the north region and had concentrated circulation in the cities within this region by 2013 and then, circulation of virus expanded to the west region and continued to 2015, therefore based on our result inferred cautiously there were small and concentrated epidemics which had occurred in most parts of the country which came along until recently, despite the rapid expansion and decline in the region, it has remained stable (Figure 1). Along with the circulation of virus within the country, the phylogeographic tree revealed that intercountry transmission occurred in circulation between southwest region, Turkey and Afghanistan and again back transmission to the north region in Iran from 2007 to 2010, in addition, same back transmission was observed in cities were located in the west region and close to the Turkey border during 2013 to 2015. So, we could infer cautiously that, the virus has entered the country many times in the form of multiple introductions and spread, and according to our data, it did not have a single independent source and the age of presence of the virus in Iran is older than neighboring countries. Based on our genome set, lineages from different parts of the country branched within the lineages from Tehran, highlighting the role of Tehran as the epicenter of the outbreaks in the country. After 2010, small and concentrated epidemics in other cities have their origins in Tehran.
According to the above-mentioned, generally we concluded that probably the virus has been circulating in Tehran with low frequency for a little period and has gradually spread throughout the region and neighboring countries in the following years in a single and sporadic cases. The small genetic distance between the different branches of these clusters in this period suggests rapid spread of the virus in this specific geographical area during the narrow time (2007-2009) in Iran and neighboring countries. Based on our genome set the virus was still present in small numbers in some countries, such as Turkey, where the virus was mainly circulating in comparison to other neighboring countries.
The H1 phylogeographic tree based on different geographical regions was constructed separately and showed that two independent clusters same as the above results were created for our total sequences and geographically suggested that the virus circulation started from north and northwest to east and southwest of Iran over time, and then moved to east and southeast in recent years (Figure 2), and this indicates the correct goodness of fit of the model.
Results of the N1 phylogeographic tree were in line with the H1 tree, which two independent clusters among sequences from Iran, Turkey, Iraq, and Turkmenistan were observed, with the sequences related to Turkey occupying the root of the tree (Figure 3). The phylogeographic tree showed that the transmission and circulation pattern was like H1 tree, and the northern region is the epicenter of this transmission which had taken place bilaterally and had occurred between the north of the country and Turkey. During a short time, the circulation of virus involved the whole country (Figure 4).
It should be noted that, if there were more sequences from other countries in our genome set, different results could be expected from the model outputs.
The evolutionary rate for the HA and NA genes of the H1N1 viruses were estimated at 4.01×10-3 [95% HPD: (2.82-5.40) ×10-3] and 3.30×10-3 [95% HPD: (1.33-5.43) ×10-3] substitutions/site/year, respectively. The mean time of the most recent common ancestor of
H1N1 viruses was 1989 (95% HPD: 1980-1994) for HA and NA as well.
The estimation of dN/dS ratio per site for each lineage was conducted. After excluding the duplicated sequences, 210 sequences with 443 sites for HA and 85 sequences with 91 sites for NA lineages were analyzed. The SLAC analysis of HA lineages showed that 17 sites were under negative/purifying selection while no positive/diversifying selection were observed. The overall dN/dS for HA lineages was 0.403. For the NA lineages, 14 negative/purifying selection sites and no positive/diversifying selection were observed. The overall dN/dS for NA lineages was 0.286. Based on the FEL analysis of HA lineages, 48 and 2 sites were under negative/purifying and positive/diversifying selections, respectively. This was 24 and 1 selection sites for the NA lineages, respectively. The results of Generic Algorithm Recombination Detection (GARD) showed no evidence of recombination.