What are cobionts?

Before we dive deep into Slimane’s fungal example, we need to understand why there might be different kinds of organisms in our sample, and how that affects the BTK plot.

When we take a sample of tissue from a large creature or a plant, for example a tiny clipping of the ear of a mammal, or a leaf from a tree, we get many millions of cells from that organism, and the sample is more often relatively free of other microorganisms. However, when we try to get DNA for sequencing from a very tiny organism such as a tiny insect or fungus invisible to the naked eye, we might inadvertently also sample all the microorganisms that live on its surface or in its gut. DNA from something other than our target organism can be found in any assembly. The most common sources are:

• Symbionts – organisms that live alongside the host organism that benefit the host (eg, Vibrio fischeri bacteria that provide bioluminescence to the Hawaiian squid Euprymna scolopes)
• Parasites – organisms that live inside or on the host, and cause damage (eg, nematode worms inside insects or plants)
• Hosts – in some cases, we might be interested in sequencing the genome of a parasite like the sheep-infecting nematode Haemonchus contortus. Even if we isolate the nematode before sequencing it, it is possible that cells from the host sheep might be in the nematode’s gut or on its skin. In this case, the contaminating organism would be the sheep.
• Lab Contaminants – common lab contaminants such as bacteria that grow on non-sterile surfaces can get into our samples of interest.
• Endosymbionts – some organisms such as insects and nematodes have endosymbiont bacteria, i.e. bacteria that live within the cells of the insect/nematode.

From the perspective of BTK, everything is a cobiont. Knowing the kind of cobiont from the list above is useful because it prepares us for understanding what we are seeing on our BTK plots. If we see a separate plant blob on our BTK plot when looking at a fungus genome assembly, it helps to know if that fungus normally infects that plant and if so, the plant cobiont blob is likely to be a real and not a mislabelled blast hit.

If we see a separate blob labelled wolbachia in a nematode genome assembly, then we can be quite confident that our nematode sample had been infected by a wolbachia endosymbiont. But if we also see a few small contigs labelled wolbachia inside the main nematode blob, then it is possible that those are wolbachia horizontal gene transfers that are present at the same concentration as the nuclear genome. Thus knowing the context of a sample matters in deciding how to identify the different organisms present in it. BTK only presents the evidence from GC, coverage, and best blast hits at different taxonomic levels. What we do with that evidence is up to us.

Once you have identified the sequences that belong to a separate organism, the next thing to decide is what to do with those sequences. If the organism is a lab / environmental contaminant, or is host DNA (eg sheep host DNA when we are interested in the sheep’s nematode parasite), then you probably just want to discard those sequences. In some other cases such as symbionts and endosymbionts you might be interested in pulling those assembled sequences (and their constituent reads) out and looking at them as separate organisms of interest. BTK allows you to do all these things, as you will see in this week’s activities.