Prof. Manuel Martinez-Garcia, associate professor of Microbiology at the University of Alicante (Spain), shared with us his experience and insight on leading the field of single-virus genomics, and the unexpected results that followed.
Scipio: « Single-cell resolution has been at the forefront of the fields of genomics for a while, but the vast majority of the studies concern mammalian systems. How did you end up trying to sequence single-viruses, with the technical challenges that it represent?”
Manuel Martinez-Garcia: « It all started during my postdoc at the Single Cell Genomics Center with my supervisor Dr. Ramunas Stepanauskas, who at the time was leading the field in single-cell genomics in prokaryotes. We were mostly working with bacteria, but in my last three months there, I was thinking that it would be cool to try with viruses. I started there and came back to Spain with the idea to further develop single-virus genomics.”
“In fact, we were not the first to use single-virus genomics technology. The novelty of our work was to apply single-virus genomics to optimize the technique and use it in a high throughput manner to discover new viruses in the environment.”
“There was a group in the Craig Venter institute that published first in 2011. We have to give them credit because they were first to try and use single-virus genomics. But in their research, they only applied the method to very well-known bacteriophages, as a proof of concept. What we did in my lab was to use that directly for environmental samples, which provide much more valuable biological insights.”
Single-virus genomics complements traditional metagenomics
S: « Shotgun sequencing techniques have been around for a while now, grabbing whichever piece of RNA or DNA you can get from the environment, and then running de novo genome assembly trying to figure out what belongs to whom. So how is single-virus genomics taking us further? »
MMG: “Metagenomics is probably one of the most powerful techniques that we have to decipher and unveil the genetic content of prokaryotes and viruses as well. But, in my case, we realized that when we try to recover the genome of the most abundant bacteria in marine systems, metagenomics was not good enough. We only got very small fragments of for example, Pelagibacter, the most abundant microbe here in the Mediterranean Sea. In reality, metagenomics is very powerful, but it’s not perfect. “
“What single-virus genomics provide, similar to single-cell genomics, is the ability to decipher and sequence every single microbe, regardless of the complexity of the system. And we think this is important because when we started sequencing marine samples, we realized that there was an abundance of viruses that were not present in the databases.”
“A good example is the Tara Oceans Expeditions, probably the biggest expedition so far for recovering shotgun sequencing data. They claimed afterward that kind of everything was discovered after this but in fact, using single-virus genomics, we were the first to discover probably the most abundant virus in our planet. Its genome was not even present in any sequencing database, which shows the limitation of shotgun sequencing when it comes to samples with high microdiversity.”
« For one of the single viruses that we discovered [using single-virus genomics] to be one of the most abundant out there, we did not find any relation in terms of genetics to the other known viruses. From the first nucleotide to the last, we did not get a match. »
A methodology for now limited to DNA viruses
S: “Talking about huge diversity, can we apply single-virus genomics to the majority of viruses, despite the large variations in sizes, capsids, or genetic material?”
MMG: “So far, the technology is not ready to capture RNA viruses. We did a project funded by the Moore Foundation to investigate, but we did not succeed because we could not find an enzyme that was sensitive enough to amplify the genetic material from a single RNA particle. I think this is the limiting step, probably the bottleneck of single virus genomics at the moment.”
“However, it works very well for virtually any DNA viral particle, as long as the genome is big enough, probably above 10 kb. There are some single-stranded DNA viruses that has very small genomes, so those ones can be problematic, but the bottleneck is really with RNA viruses.”
Viral taxonomy has blurry lines, complicating the characterization of “new” strains or species
S: “So if you amplify thousands of viral genomes, how are you able to quantify the existence of different strains, or even species? Is there a threshold for the minimal number of copies of a specific genome sequence before it can be inferred to be a stable variant or subspecies?”
MMG: “Well, this is a broader question than you think, because we are only barely starting to use standardized language. What I think is a viral species probably is not the same as other scientists. Viral taxonomy is very complex, I am not sure if viral species truly exist. When I discovered using single-virus genomics a new virus, or at least what I call new, maybe it is not so new for others, or maybe it is too new for me.”
“Similar to other techniques, you need to compare your ”new” genome to existing databases, and the degree of relatedness against the other genomes is what gives you the information you need. New is only an adjective, and it is very subjective.”
“For example, for one of the single viruses that we discovered to be one of the most abundant out there, we did not find any relation in terms of genetics to the other known viruses. From the first nucleotide to the last, we did not get a match.”
Single-virus genomics brings new insights into the diversity of viral strains
S: “Putting the definition of a “new” species aside, what were your unsuspected discoveries using your high-throughput single-virus technique?”
MMG: “One of the surprises with single-virus genomics was realizing that the diversity is huge, and the micro diversity is even higher than we expected. Our initial idea was that within each one of the species we would have a few strains but in reality, we found out that there are thousands of different coexisting strains belonging to the same species, always present in a sample.”
“All these strains are a continuous line of diversity: it is not two or three distinct variants, but thousands sharing 80% to 100% similarity and coexisting. It is almost overwhelming. We found this huge micro-diversity in every sample we’ve been analyzing, from the Mediterranean, but also the Atlantic and Arctic. It completely blurs the lines, there are no clear separations. It makes it really hard to put variation thresholds for variants or species, or even discussing and agreeing on such thresholds.”
These results are described in detail in an upcoming paper by Prof. Manuel Martinez Garcia and his team. Regarding the technical approach to single-virus genomics, we will be sharing Prof. Manuel Martinez Garcia’s advice soon in another dedicated article.
- Martínez Martínez J, Martinez-Hernandez F, Martinez-Garcia M. Single-virus genomics and beyond. Nat Rev Microbiol. 2020 Oct 6. doi: 10.1038/s41579-020-00444-0.
- Garcia-Heredia I, Bhattacharjee AS, Fornas O, Gomez ML, Martínez JM, Martinez-Garcia M. Benchmarking of single-virus genomics: a new tool for uncovering the virosphere. Environ Microbiol. 2021 Mar;23(3):1584-1593. doi: 10.1111/1462-2920.15375.