Jacob Mayfield Aug 08, 2014

As the worst Ebola outbreak in recorded history continues to devastate Africa, it's easy to see why viruses have a reputation as villainous, sinister, robot-like killers.  But just as the mantra "the only good microbe is a dead microbe" has given way to FDA guidance on fecal transplants, so too should our fear of viruses be tempered by an understanding of their role in the holobiont.


The past few years have seen the beginnings of that transition with a few published studies on the viral components of the human microbiome.  My favorite is Forest Rohwer's discovery that not only are human mucosal membranes colonized by specific bacteriophage, the phage may play a protective role lysing invading bacteria.  My suspicion has long been that some of the viral remnants in our own genome influence our viral microbiome, perhaps by modulating bacteriophage that sebsequently influence our bacterial friends and foes.  In the mean time, given the enormous role viruses play in global microbiology (as the most abundant life form on earth and responsible for 10-20% of bacterial deaths every day in marine bacteria), why don't we know more about them?


Viruses defy the central dogma of molecular biology (information flows from DNA to RNA to protein.  Not! See RNA viruses), resist phylogenetic characterization (spiting Carl Woese by not bringing their own SSU rDNA), and by encoding a dizzying array of unique proteins (leading Graham Hatfull to refer to "their pervasively mosaic nature").  In short, we know disproportionately little about [nonpathogenic, nonmodel] viruses because our genomic pipelines aren't very good at finding and characterizing them.  We have to look specifically for viruses, isolating viral particles, checking for genomes in the RNA pool, or changing our genome assembly pipeline.  Recognizing this, the NIH funded Human Microbiome Project is funding human virome projects, like this one on human gut virome diversity.


In the mean time, I was excited to read about a virus that most of us have but didn't even know existed.  Using a technique called cross assembly that looks at correlations in read depth, Bas Dutilh and colleagues found a phage genome present in the vast majority of human microbiome metagenomes that had previously been thrown out with the bathwater.


Its abundance is not surprising considering is appears to infect Bacterioides, a major bacterial clade in the human gut resposible for over a billion cells per gram of feces.  We still don't know much about the phage and the authors were unable to culture it.  But I'd like to give it a big, symbiotic welcome.