Log in

Sun, Nov. 27th, 2005, 10:51 am
Bio experiments

There's an open question as to where viruses originally came from. Did they coevolve with bacteria, or are modern viruses things which spontaneously came into existence in the recent past?

Here's an interesting experiment which might shed some light on the issue: Create an environment which is very amenable to bacterial growth, meaning plenting of oxygen, water, heat, and nutrients, but in which there are no life forms whatsoever. Then introduce a single bacterium, and let it grow for many generations. At the end, check to see if any viruses have appeared.

And here's an unrelated experiment in psychology. Let us say that we wish to measure just how bad a bias an experiment not being double-blind would create. For the purposes of this experiment, we ask for volunteers to be both testers and testees in a sham experiment having to do with subliminal messaging. The subject is shown a flash of something, and the tester is shown what it was but the testee is not. Of course, there isn't ever any actual subliminal message shown - the experiment is to see how much the testee guesses correlate with what the tester thinks the testee saw.

Such meta-testing is done quite rarely. Whether males and females interpret the meanings of numbers in a 1 to 5 scale is, so far as I know, a completely untested area, making a very large field of survey-based gender psychology extremely dubious. The book Monkeyluv, which I highly recommend, has a section talking about a meta-study of controlled rat experiments where multiple groups attempted to raised rats under exactly identical conditions and then measure various standard traits about them, and the results indicate that many important traits, including some which various drugs are prescribed to control, are mostly controlled environmentally. The study's implications have unfortunately been mostly ignored.

Sun, Nov. 27th, 2005 09:46 pm (UTC)

I think you'd be unlikely to see viruses evolve without an environment for them to evolve in. IANAB (I Am Not A Biologist), but viruses lack most of the structures that allow an organism to be classified as 'life' - an (immediate) way to reproduce, metabolise food, etc. They hijack other cells and 'modify' them to perform reproduction task. Thus, in your experiment there'd be no way for viruses to evolve, since the environment lacks sufficiently sophisticated cells for the virus to invade.

I'm assuming that viruses can't invade bacteria; from some cursory digging around, I can't find a reason why this wouldn't be possible. If they can, I'd expect to see a race of reproduction rates between bacteria and viruses. I doubt that the new viruses be able to out-reproduce the bacteria sufficiently for us to notice them, particularly given that the bacteria are already more evolutionarily successful than the brand new viruses. If they could 'take root' in the bacteria and achieve an infection rate greater than the bacteria's reproduction rate, you'd see a whole lot of bacteria dying off. This would be quite visible to us, since by now you'd be talking about a giant smelly petri tank full of bacteria.

I suspect that the viruses wouldn't appear in the first place. There'd be no evolutionary pressure against death (due to the favourable environment) and pressure towards pure reproduction rates (increased metabolic rates, faster cell division). This goes counter to viruses, which perform neither metabolism nor division.

Of course, an anti-bacteria-specific virus may eventuate (and would be likely the only virus to spontaneously evolve in this environment). This requires my first (second?) assumption, that bacteria are in fact vulnerable to the mechanism that we use to classify an entity as a virus. Is this abstract enough yet? :-)

As for meta-testing in psych, I'm not aware of any studies, but there is an awareness of the limitations of an arbitrary numerical scale (IANAP, but I used to date one. Therefore, I'm an expert!) This has led to experiments using things like "1 2 4 5" scales, where the test subject cannot choose the neutral choice - they must swing in one direction or the other.

Sun, Nov. 27th, 2005 10:31 pm (UTC)

I'm afraid most of what you said about viruses and bacteria is wrong, there are many viruses whose only host are bacteria and many of these are very important in their uses in experimental biology. There's also no reason that there has to complete dominance of viruses or bacteria, after all we don't see that as the case in natural populations except where one species has an evolutionary advantage. In these cases we get either an epidemic or the extinction of the virus. In fact most viruses are maintained at a balanced level in a population which is an evolutionary advantage for them.

There are also many techniques available to us that let us visulise small amounts of cell death and to determine its cause. This sort of experiment is very interesting though, however it would require a complex enviroment and a very large scale in order to find results after all it has taken millions of years for viruses to evolve to their current complexity.

Mon, Nov. 28th, 2005 09:44 am (UTC)

Ooh - someone who sounds like they know what they're talking about! :-)

Knowing that there are viruses that invade bacteria makes things simpler in my mind ('my mind' being headachey and missing memories after being thwacked on the ground in a mountain biking crash a few hours ago, so if this makes absolutely no sense, I'd appreciate it if you'd let me know!)

In the experimental conditions, I'd expect dominance of one or the other (with some caveats). This would be dependant on the infection rates of the virus vs. the growth rates of the bacteria. I'd expect the bacteria to grow rapidly and only die off when they exhaust their food. Assuming an infinite food supply and the presence of viruses which can infect bacteria:
- if the virus infection rate < bacteria growth rate, both will grow continually, but the number of viruses in the system will be very small compared to the number of bacteria
- if the virus infection rate > bacteria growth rate, eventually all bacteria will be infected. This will inhibit their growth rate and eventually lead to the extinction of the bacteria (or the evolution of a new breed of bacteria which is resistant or immune to the virus).

Your point about restricted virus infection rates being an evolutionary advantage is interesting; it is true, but I don't think it applies here. We're talking about an isolated system. If the viruses managed to completely overwhelm the bacteria, they'd die off and be an evolutionary failure. If you set up *many* discrete systems, you might eventually encounter one where the virii and bacteria co-exist with relative stability; these would be evolutionarily successful viruses. Assuming that the bacteria didn't evolve to resist the viruses, or the viruses didn't 'get greedy' and evolve a way to infect faster (which would be an evolutionary advantage in the short term).

I agree on the timescale point, though - I doubt we'd see viruses spontaneously appear within our lifetimes.

Mon, Nov. 28th, 2005 11:06 am (UTC)

bacterial viruses come in two main types - bacteriophages and plasmids.

a successful parasite of any kind will infect it's host population in such a way as to ensure it's own replication without destroying the host population and thereby removing it's future chances of replication.

this dynamic is the same for predator-prey populations as well, thought he balances will differ.

the rate fo bacterial evolution is high due to features of bacterial genetics, and their generally rapid replication rates. their viral populations tend to be similarly adaptive. the original arms race, if you like :)

plasmids are pieces of infectious dna that can be passed between bacteria. they tend to offer a 'bonus' to their host in return for replication and being passed on. for example, most bacterial resistance to antibiotics is due to plasmids.

it would be hard to know if a human virus "spontaneously" appeared. they are pretty complex little darlings, and usually evolve from other human viruses, or occasionally either manage to adapt to humans or happen to be able to infect us.

the new guys are the ones that kill and cause serious problems. neither we nor they have adapted to co-exist properly. over time, a successful host-parasite relationship will develop - or one will destroy the other.

standard population genetics.

Mon, Nov. 28th, 2005 10:41 am (UTC)

we were pretty sure viruses had been around for a very, very long time.

human-specific viruses could only be as old - or younger - as the human species.

as for the details of how particular varieties of viruses evolved, that's part of the fun of virology and molecular genetics :)

Sun, Dec. 11th, 2005 07:54 am (UTC)

Presumably viruses which were around at the dawn of humanity were the same ones as in the immediate pre-humans. The line in the sand doesn't mean much.

Does genetic clock work well on viruses? It's hard to see how it could tell you anything at all about viroids, since those have hardly any genetic material to speak of.

Mon, Dec. 12th, 2005 01:30 am (UTC)

viruses mutate/change faster than more complex organisms. there's lees of them, so any change has a statistically greater probability of having an impact.

it is possible to map out linkages between populations of viruses. being able to isolate them from ancient, preserved bodies/carcasses is also quite helpful.

it is only possible for a virus or other parasite to be species-specific after the population of variants settles down into a stable species. to be host specific, requires a close association with the host, and that association would include accessing mechanisms (usually cellular entry) specific to say, humans.

because there are so many commonalities between the cellular proteins of related, and near-related, species that virusescan 'jump' specied so readily. they have to have both access to the new host - and a way into the hosts system. the influenza family get that through subsistence farming and/or poor hygine - and there are sufficient similarities between the avian and mammalian cell membrane proteins they latch onto to gain entry.

it is possible, through diligent work, to establish family trees for viruses, particularly the more complex vertabrate ones. the h.i.v. genome is quite complex. it easily supports a huge variety of changes without losing effectiveness. statistics (as loved by epidemiologists) coupled with genome or proteome maps enables one ot establish links. enough of those gives you a picture of the evolution - and hence the procession - of the 'entity'.

i do miss viriology.