NEW YORK (GenomeWeb) – A new study of the tardigrade genome calls into question a titillating, widely reported paper which suggested that genome exhibited high levels of foreign horizontal gene transfer (fHGT).
In November, scientists led by Thomas Boothby and Bob Goldstein of the University of North Carolina at Chapel Hill published a study in the Proceedings of the National Academy of Sciences claiming evidence of "extensive" foreign DNA present in the genome of the tardigrade species Hypsibius dujardini.
Now, a group of British scientists led by Georgios Koutsovoulos and Mark Blaxter of the University of Edinburgh have performed their own analysis of the H. dujardini genome and concluded that the previously reported findings were primarily due to bacterial and other contamination.
"Tardigrades are amazing organisms, but these suggestions about their DNA were a step too far," Blaxter said in a statement.
In the paper published this week, also in PNAS, he and his colleagues said the UNC-led group's genome was "poorly assembled and highly contaminated." At press time, Boothby and Goldstein had not responded to GenomeWeb's request for comment.
Tardigrades are microscopic animals renowned for their ability to survive in extreme environments. Boothby and Goldstein performed their original study in search of the mechanisms that allow tardigrades to survive extreme stresses. Preparing and analyzing a draft genome assembled using a combination of short-read sequencing data from Illumina and long-read data from Pacific Biosciences and Moleculo, they proclaimed that more than 6,600 genes, one sixth of the H. dujardini genome, were obtained via fHGT.
The new study puts forth its own draft genome and analysis based on Illumina short-read technology, casting doubt on Boothby and Goldstein's conclusions. H. dujardini displays evidence of fHGT, the authors said, but on the scale of 1 percent of the entire genome, comparable with the estimates for well-established model organism genomes such as fruit flies and primates.
"Overall, very few of the total fHGT candidates proposed by Boothby et al. were in scaffolds that were not obviously contaminant," the authors wrote. "Genomic sequencing of small target organisms requires the pooling of many individuals, and thus also of their associated microbiota, including gut, adherent, and infectious organisms. Contaminants negatively affect assembly in a number of ways and generate scaffolds that compromise downstream analyses."
The Edinburgh-led group used the same stock culture of H. dujardini to prepare its genome assembly. In addition to carefully cleaning the samples prior to DNA extraction, the scientists also employed bioinformatics-based techniques to remove contamination after preliminary assembly.
"Read pairs contributing to contigs with bacterial identification and no mitigating evidence of tardigrade-like properties — GC percent, read coverage, and association with eukaryote-like sequences — were conservatively removed," they said.
While the UNC-Chapel Hill assembly measured a span of 252 megabasess, the Edinburgh assembly measured 135 megabases. Analyzing the UNC assembly, the Edinburgh researchers said about a third of it appeared to derive from contamination by bacteria, including Armatimonadetes, Bacteroidetes, Chloroflexi, Planctomycetes, Proteobacteria, and Verrucomicrobia; rotifers, including Adineta vaga; and mimiviruses.
Blaxter's group first announced their results Dec. 1, 2015, on the preprint server BioRxiv.
"What would in decades past have taken many months to sort out became the focus of experts around the world and has been swiftly resolved," Blaxter said. "We hope this paper will finally correct the scientific record."