Researchers in Italy have retracted a 2019 paper on the genetics of a form of herpes virus after determining that the genomic sequences they thought they’d been analyzing proved to be something else.
The paper, “A complex evolutionary relationship between HHV-6A and HHV-6B,” appeared in July in Virus Evolution, an Oxford University Press title. The authors came from the Scientific Institute IRCCS in Lecco, and the University of Milan.
The article purported to find that:
Recombination detection methods identified these viruses as interspecies recombinants, but phylogenetic reconstruction indicated that the recombination signals are due to shared ancestry. In analogy to iciHHV-6A, NY cluster viruses [obtained from samples in New York state] have high nucleotide diversity and constant population size. We propose that HHV-6A sequences and the NY cluster population diverged from an ancestral HHV-6A-like population. A relatively recent bottleneck of the NY (or a related) population with subsequent expansion originated most HHV-6B genomes currently sampled. Our findings indicate that the distinction between HHV-6A and -6B is not as clear-cut as previously thought. More generally, epidemiological and clinical surveys would benefit from taking HHV-6 genetic diversity into account.
But as the researchers learned, the viral genomes they’d been using for their study were incorrect, thanks, they said, to errors in the GenBank repository. According to the retraction notice:
The authors of “A complex evolutionary relationship between HHV-6A and HHV-6B” (doi: 10.1093/ve/vez043) have requested retraction of their paper. All authors agree with this action. Virus Evolution is retracting the paper because its findings are unreliable due to honest experimental error.
Specifically, analyses and conclusions were based on the retrieval of genome sequences from GenBank that were erroneous (accession numbers: KY239023.1, KY274487.1–KY274525.1, KY290171.1–KY290221.1, KY315520.1–KY315558.1). These genomes have been since corrected on GenBank (accession numbers: KY239023.2, KY274487.2–KY274525.2, KY290171.2–KY290221.2, KY315520.2–KY315558.2). Raw read data for the genome assemblies are available on SRA (accession numbers: SRR8717229, SRR8745772–SRR8745811, SRR8749118–SRR8749151, SRR8749387–SRR8749420, SRR8757190–SRR8757227, SRR9736967–SRR9737010).
Since this mistake affects a significant portion of the paper’s results the journal has decided that a retraction, rather than a correction, is the most appropriate course of action. Virus Evolution apologises for any inconvenience caused by this retraction, and thanks the authors for their rapid action and open communication in dealing with this matter.
The GenBank entry for the first sequence, KY239023.1, notes that it was submitted in April 2017 and replaced in October 2019, several months after the Italian researchers published their paper. Same for KY274487.1, KY274525.1 and the others that we checked.
Oliver Pybus, the editor of Virus Evolution, praised both the authors and the researchers who generated the original data and then corrected the sequences:
[T]heir openness and responsiveness was commendable.
GenBank seqs do occasionally get updated. But it’s perhaps unusual that the seq changes led to such substantial alterations to the conclusions of the paper.
My take is that the benefits of genomic data sharing vastly outweigh the occasional costs, such as this. Everyone acted in good faith and the system self corrected quickly.
The corresponding author of the paper has not responded to a request for comment.
Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at [email protected].