Re: A question about recombination (could contamination be an issue?)
Here are a couple of comments on the issue of lab contamination. The one set of sequences that was noted in the commentary
http://www.recombinomics.com/News/10120601/H5N1_China_Evolution.html
had obvious recombination that was noted by the submitters, Beijing Genome Institute, in the title of the submission:
“A cohort of AIV H5N1 subtypes isolated from wild aquatic birds and domestic poultry revealed rapid transmission, frequent reassortment, and
identifiable recombination events ” (emphasis added)
The "identifiable" recombination was quite obvious. These sequences were submitted on February 28, 2005 and released on February 28, 2006. The one set of sequences was A/
chicken/Henan/210/2004. The same group released the same set of sequences in September, 2006, but used letters instead of numbers for the sample. Thus, the species, location, and date were the same, but the sample number was changed from a number to letters, so sample 210 above became wu as in A/
chicken/Henan/wu/2004. The sequences of each of teh 8 gene segments from these two submissions were either identical, or differed by a single nucleotide, leaving no doubt that the second submission was a resequencing of the first submission.
The names on the two sets of submiters were similar and the submissions were from the same institute. All of the sequences from 2004 were resubmitted in 2006, which was obvious when sequences were compared (the sequences clearly matched and the differences were sample numbers - the later submissions all used letters, while the earlier submissions all used numbers).
However, although all of the above is quite obvious, the issue of contamination is addressed equally well by either scenario. If the isolate from the same sample was resequenced 1 1/2 years later and gave the same result, then the recombination was not due to contamination.
If all of the later submissions coincidentally all came from the same species from the same location, on the same date, and coincidentally all matched, then again the recombination was not due to contamination, because the same recombination was seen in both sets of sequences.
Thus, regardless of whether the identical recombination result came from sequencing of two isolates from the same sample or came from two isolates from idependent samples that coincidentally matched the species, location, and date (multiple times), the bottom line was the same - the reproduced recombination in the two sets of sequences was NOT due to contamination.
