My thoughts on an Evolution News post on ERVs

[Rumraket (Mikkel R.)]

@scd

again, we can see the opposite too (a part of a retrovirus that was created from host genes).

Reference please.

[scd]

ok guys. i feel that we just walking in circles so lets make it simple: lets take 2 claim and discuss about them. in this way it will be easy to see who is right. so my main claim is that retroviruses can evolved (actually “can be created” should be more precise description since no new complex biological system evolved here. just moving one complex part to other place). one reference for this claim is the V-Src gene that was taken from the host by a retrovirus:

v-Src

so we have evidence that viral parts can be created from the host genome. here is another example and i can bring you more:

Retroviruses from retrotransposons

bottom line- this is possible explanation. believe it or not.

my second claim is that it might be possible to get many identical insertions at the same spots. again, we have evidence for that claim too (although they are talking about introns the odds should be similar to the erv case):

Introns -- nonsense DNA -- may be more important to evolution of genomes than...

“Remarkably, we have found many cases of parallel intron gains at essentially the same sites in independent genotypes,” Lynch said. “This strongly argues against the common assumption that when two species share introns at the same site, it is always due to inheritance from a common ancestor.”

so as we can see, we have evidence for both possibilities.

[John_Harshman]

Sorry, but your example doesn’t show what you claim. This isn’t a new virus arising from host DNA; it’s a previously existing virus incorporating some host DNA, something that happens all the time, and that DNA being retained because it’s advantageous to the virus. What you have here is just mutation and selection, not creation and not even a brand-new virus.

Your other example (as you even point out) is not relevant to retroviral insertions, unless the introns are themselves retroviral insertions and “essentially the same sites” actually means “at the same sites”.

[T_aquaticus]

@scd

so my main claim is that retroviruses can evolved (actually “can be created” should be more precise description since no new complex biological system evolved here. just moving one complex part to other place).

We observe ERVs being created by the insertion of exogenous retroviruses.

We do NOT observe supernatural deities creating genomes with fake retroviral insertions in them.

When we have an observed natural process that can produce the phenomenon under question we go with the natural explanation. It’s called parsimony.

@scd

so we have evidence that viral parts can be created from the host genome.

Retroviruses can insert into a host genome, and those parts can evolve other functions over time. Evolution of ERVs after insertion does not cast doubt on their origin.

@scd

my second claim is that it might be possible to get many identical insertions at the same spots.

You need to show how we can get 99.9% of 200,000 viral insertions to occur at the same base in two different genomes. Not many, but nearly all. Where is the evidence for this?

More to the point, we can observe that retroviruses only very rarely insert into the same base. Again, THESE ARE OBSERVATIONS. For example:

Relationship between Integration Sites and Transcriptional Intensity in the Human Genome

The human chromosomes are shown numbered. HIV integration sites from all datasets in Table 1 are shown as blue “lollipops”; MLV integration sites are shown in lavender; and ASLV integration sites are shown in green. Transcriptional activity is shown by the red shading on each of the chromosomes (derived from quantification of nonnormalized EST libraries, see text). Centromeres, which are mostly unsequenced, are shown as grey rectangles.
Retroviral DNA Integration: ASLV, HIV, and MLV Show Distinct Target Site Preferences - PMC

3 different retroviruses inserted along the entire length of all 46 chromosomes.

[scd]

@John_Harshman

This isn’t a new virus arising from host DNA;

@T_aquaticus

We do NOT observe supernatural deities creating genomes with fake retroviral insertions in them.

i showed that host parts (such as retrotransposon) can become a retrovirus. if a part of a retrovirus can be created from host parts then why not the whole retrovirus?

@John_Harshman

Your other example (as you even point out) is not relevant to retroviral insertions

why not? in both cases we are talking about random insertions.

@T_aquaticus

More to the point, we can observe that retroviruses only very rarely insert into the same base. Again, THESE ARE OBSERVATIONS

you forgot natural selection. if the whole genome is functional (as some scientists claims) then there are only specific spots in the genome that we can get viral insertion without any problematic result to the creature. thus if this scenario is true we should find viral insertions only in specific spots.

[Roy (34-year veteran)]

Illustrating the Fallacy of Composition:

@scd

i showed that host parts (such as retrotransposon) can become a retrovirus. if a part of a retrovirus can be created from host parts then why not the whole retrovirus?

If part of a car can be made out of rubber, why can’t the whole car?

[Dan_Eastwood]

@scd

i showed that host parts (such as retrotransposon) can become a retrovirus. if a part of a retrovirus can be created from host parts then why not the whole retrovirus?

Are you saying that such thing might evolve naturally? By simply random chance?? Welcome to the evolution club!

[Rumraket (Mikkel R.)]

@scd

bottom line- this is possible explanation. believe it or not.

You’re confused. Nobody on this side of the debate says bona fide viruses can’t evolve from host genes.

The discussion here is about which is the most plausible explanation for the existence, distribution, and nature of ERV sequences in human and other primate genomes, which can be used to infer their shared genealogical relationships.

You are attempting to argue that ERV insertions in common between humans and other primates, and their nesting hierarchical structures, can be equally well explained by these species having been independently created to carry these ERV genomes for functional reasons, and that instead of them resulting from ancestral retrovirus infections(to explain their similarity to retroviruses), these sequences evolved into extant retroviruses.

But the problem is this doesn’t explain their distribution (why they reflect the canonical phylogeny of primates). While that may explain the mere fact of their similarity to retroviruses(which the hypothesis of insertion of actual ERVs also does), it does not explain their sequence-based phylogenies or their mutual consilience.

You are attempting to rebut the inference that they originated as insertions by retroviruses by showing that such insertions can evolve back into retroviruses. But that doesn’t explain why they are present in the first place or why they have the sequences they do. You seem to only explain the existence of retroviruses (they exist because they evolved from ERV-like sequences in host genomes), not why the ERV-like insertion is there in the first place or what function their similarity to retroviruses serve.

None of the hypotheses you have waved your hands towards to argue that ERV-like sequences could be functional, require them to take the form of retrovirus-like genome sequences in order to perform their functions as putative transcriptional regulators.

Also worthy of consideration here is that retrovirus insertions by retroviruses are much more frequent than ERV sequences evolving into retroviruses. That is after all the basis for every single retroviral infection that ever occurred.

[Giltil (Gilbert Thill)]

@Roy

If part of a car can be made out of rubber, why can’t the whole car?

Isn’t it the same fallacy used by evolutionists to explain IR systems: if a part of an IR system can arise by RV+NS, why not the whole IR system?

[Rumraket (Mikkel R.)]

For the record I don’t think what he said was actually fallacious. I just think the scenario he describes is less likely than the one he is trying to replace. His hypothetical appears to be an attempt at explaining the existence of retroviruses by the more or less chance combination and evolution of disparate genetic elements in some host organism, ultimately leading to the formation of some sort of ERV-type “insertion”, that then takes final plunge and evolves into a fully-fledged retrovirus. Can that happen? Sure, I think that in principle it can.

But it seems to me that scenario is much less probable than an ERV insertion simply deriving from an already existing retrovirus.

[Roy (34-year veteran)]

No.

[John_Harshman]

@scd

i showed that host parts (such as retrotransposon) can become a retrovirus.

No you haven’t. You’ve referenced cases in which a retroviral insertion, not host DNA, can pick up a piece of host DNA when it’s transcribed and carry that away with it. This is not assembly of a new virus from host parts.

@scd

in both cases we are talking about random insertions.

Are they?

@scd

if the whole genome is functional (as some scientists claims) then there are only specific spots in the genome that we can get viral insertion without any problematic result to the creature.

But that isn’t true. VIral insertions happen all over genomes. They just happen more often in more frequently exposed bits. And some scientists are quite wrong.

[scd]

@Dan_Eastwood

Are you saying that such thing might evolve naturally?

remember that we are talking about moving existing complex structure to other place. not a new complex structure. so i will not call it “evolution”

@Rumraket

But the problem is this doesn’t explain their distribution (why they reflect the canonical phylogeny of primates

are you asking why they show the same phylogeny as any tipical gene? if so the answer is simple: these “ERVs” arent so different from any other gene.

@Rumraket

You seem to only explain the existence of retroviruses (they exist because they evolved from ERV-like sequences in host genomes), not why the ERV-like insertion is there in the first place or what function their similarity to retroviruses serve.

see above: ERV can be integral part of the original genome.

@John_Harshman

You’ve referenced cases in which a retroviral insertion, not host DNA, can pick up a piece of host DNA when it’s transcribed and carry that away with it

how does it matter? in any case a host part becoming a retrovirus: “It appears that the transition from nonviral retrotransposon to retrovirus has occurred independently at least eight times,”

@John_Harshman

And some scientists are quite wrong.

you should tell them.

[Rumraket (Mikkel R.)]

@scd

are you asking why they show the same phylogeny as any tipical gene? if so the answer is simple: these “ERVs” arent so different from any other gene.

That’s not an explanation at all. ERVs are different from innumerable other genes, which is how they are even recognized as ERVs: Their similarity to retroviruses, and their dissimilarity to things that aren’t retroviruses. But none of this has to do with the consilience of the phylogenies derived from ERVs and other genes, as two very similar genes and two very dissimilar genes can still produce identical phylogenies.

Your answer literally makes no logical sense.

@scd

see above: ERV can be integral part of the original genome.

That does not explain why they need to look like ERVs. A transcription factor binding site doesn’t need to look like, much less take the size of a typical LTR. LTRs are typically several hundred basepairs. Transcription factor binding sites are typically 6-30 bases, with most in the 10-bp range.

There’s no reason why you’d need two LTRs flanking GAG, POL, and ENV, as in several tens of thousands of basepairs, just so you can get a handful of transcription factors to bind and initiate long-range regulation. Their presense, sequence, and synteny only make sense if they derive from retroviruses. None of these attributes are necessary for whatever real or imagined cellular functions you’ve so far referred to.

[John_Harshman]

@scd

in any case a host part becoming a retrovirus: “It appears that the transition from nonviral retrotransposon to retrovirus has occurred independently at least eight times,”

Where does the gene that turned the retroelement into a virus come from, according to that paper? Hint: it’s not from the host genome.

[scd]

@Rumraket

ERVs are different from innumerable other genes, which is how they are even recognized as ERVs

its also true for retrotransposons and transposons. and they are probably not the result of viral insertion.

@Rumraket

That does not explain why they need to look like ERVs. A transcription factor binding site doesn’t need to look like, much less take the size of a typical LTR.

@Rumraket

There’s no reason why you’d need two LTRs flanking GAG, POL, and ENV, as in several tens of thousands of basepairs,

see above: its also true for other (functional) host components. i also showed here evidence for convergent insertions. if you think that its impossible to get about 100,000 insertions at the same spots please show your calculation.

@John_Harshman

Where does the gene that turned the retroelement into a virus come from, according to that paper? Hint: it’s not from the host genome.

this is another case of circular reasoning. they assume that this part has a viral origin. when in reality its possible that this part is actually an integral part of the host genome.

[Rumraket (Mikkel R.)]

You’ve stopped making any sense I’m sorry to say.

[scd]

can you be specific where?

[John_Harshman]

“Stopped”?

[T_aquaticus]

@scd

i showed that host parts (such as retrotransposon) can become a retrovirus. if a part of a retrovirus can be created from host parts then why not the whole retrovirus?

That’s part of the retroviral life scycle. It inserts into the host genome, and then translated copies of the insertion become viral particles. All you are doing is pointing to the natural and well explained workings of retroviruses. This in no way puts the natural retroviral origin of ERVs into doubt.

@scd

if the whole genome is functional (as some scientists claims) then there are only specific spots in the genome that we can get viral insertion without any problematic result to the creature. thus if this scenario is true we should find viral insertions only in specific spots.

You need evidence, not bare assertions.