Commentary
Recombination Confirmed in Canadian Swine Influenza
Recombinomics Commentary 17:04
June 29, 2008
There is a highly significant overlap between the sequences showing evidence of apparent homologous recombination listed in Appendix 2, and the sequences showing evidence of apparent “frozen evolution” listed in Appendix 1. Similarly to the “frozen” viruses, isolates often have apparent evidence of homologous recombination in more than one segment, and the overall incidence of recombinant pairs is consistent across the 8 segments, allowing for greater statistical power to resolve such pairs in the longer segments.
Along the same lines, nearly all anomalous sequences come from avian and swine hosts; it seems natural to assume that viruses from human hosts are generally handled with greater care (because of the potential public health hazards resulting from their spread) and are thus less susceptible to stock contamination.
The above comments from were used to argue that the clear cut homologous recombination discussed in the paper “Anomalies in influenza genome database: New biology or lab errors?” represented in Appendix 2 were due to lab error. However, one series of human HA sequences had clear cut recombination in six isolates, which fell into two major groups. These groups were internally consistent, strongly suggesting that they were not due to lab contamination by stock viruses. In this case the stock virus would have been one in circulation a decade earlier. The data against such contamination was not only provided by the sequences deposited by the Korean CDC in 2004, but was also provided by sequences generated independently in the US and Japan and deposited earlier this year. The re-sequencing generated exact matches.
However, these human HA H3N2 sequences also demonstrated multiple cross-overs in one of the two series. These sequences demonstrated how longer stretches of sequences acquired from one parental sequence, could be cut down to small segments via additional recombination events. These smaller segments indicate homologous recombination is common, as are smaller segments, which is what was found when human sequences were analyzed in ,“Homologous recombination is very rare or absent in human influenza A virus”, where the likelihood that the result was due to chance was less than 1 in a billion.
Thus, the human data provides compelling evidence for recombination, as well as multiple cross-overs, resulting in short stretches of acquired sequences.
However, the same arguments can be used in the Canadian swine sequences, which are included in Appendix 1 and 2 in the paper quoted above. Although the Canadian sequences have not been independently confirmed, they do have evidence for internal constancy, as well as nesting of acquired sequences which also reduce the size of the stretches of identity with earlier isolates. This is demonstrated most conclusively in the PB2 sequences. Sequences from isolates 57561 and 56626 have acquired the exact same sequence, beginning at position 550 and ending at 1594, providing internal consistency. However, the acquired sequence is composed of sequences that match three other isolates. The sequences between 550 and 755 match 11112, and then switch to a 1998 North Carolina sequence, which extends to position 1594 for all three Canadian swine sequences (57561, 56626, 11112). However, 57561 and 56626 have a 1977 sequence from Tennessee nested between positions 1006 and 1326. Thus, this acquired sequence is broken down into short alternating sequence with crossover points at positions 550, 755, 1006, and 1326.
These identical cross-over points in two or three isolates are more rigorous confirmation that the recombination is due to natural recombination events and not due to lab contamination, which would produce different cross-over points in different sequences.
Thus, the Canadian swine sequences listed in Appendix 1 and 2, as well as the series of human sequences from Korea, demonstrate clear cut homologous recombination involving faithfully replicated sequences found in isolates collected 1 to 3 decades prior to the generation of the contemporary sequences.
These data provide compelling evidence that the homologous recombination in human and swine sequences are not due to lab contamination. These examples involve significant differences between the two sets of parental sequences. However, frequent recombination between closely related sequences would lead to frequent and predictable acquisitions of single nucleotide polymorphisms.
Media Links
Recombinomics Presentations
Recombinomics Publications
Recombinomics Paper at Nature Precedings