That makes no sense to me since they are included here as LTRs, notice it includes ERVs class I, II, III and others.
Retroelements and the human genome: New perspectives on an old relation
As I said, ERVs are sometimes distinguished from solo LTRs, rather than referring to both as ERVs. The operative word is 'sometimes.' On the same page that contains figure 1, the authors refer to them as "LTR-containing retroelements (
Bannert & Kurth, 2004, p.14572)." Solo LTRs can be referred to as ERVs (where the other type is reffered to as full-length ERVs), or as only that which an ERV contains. Hence the confusion, and the reason for emailing the authors of
Polavarapu et al. (2006).
That is not what you said, this is what you said:
When we examine the collective genome of Homo sapiens, we find that a portion of it consists of ERVs (IHGS Consortium, 2001). We also find that humans share most of them with Chimpanzees
We find that we share 'most of them' and yet the most abundant ERVs in the Chimpanzee genome are absent in ours.
I am referring to the set of solo LTRs and full-length ERVs, together, as 'ERVs.' Again, the authors of
Polavarapu et al. (2006) may not have been doing so when they said PtERV1 is "one of the most abundant families of endogenous retroviruses in the chimpanzee genome."
And again, even if they were, they would simply be wrong, as the genomes of chimpanzees and humans contain tens of thousands of ERVs, i.e. solo LTRs + full-length ERVs. I do not know where we are misunderstanding one another, here. Is it that you are claiming that chimpanzees and humans do not posses tens of thousands of ERVs? If you are not claiming this, since you know PtERV copy number is only in the hundreds, you must know that a couple hundred is only a small fraction of tens of thousands. Most ERVs are shared, and PtERV1 represents much of the small fraction that are not shared. They may be "one of the most abundant," considering only full-length ERVs, but ~200 simply isn't the majority of tens of thousands.
Feel free to email the authors and while your at it maybe you want to email the Chimpanzee Genome Consortium.
You telling me to email the
CSA Consortium (2005) authors, yet I just told you that I did so. I even quoted the email I got from Dr. Waterston. You even quoted in full the portion where I explained all this and where I quoted him.
And his response was to confirm what I explained to you just now. Here it is again:
In
CSAC (2005), table 2 lists the only lineage-specific, i.e. non-orthologous ERVs the researchers could find in each lineage, of the tens of thousands present. To corroborate that this was the case, I emailed the authors, and they confirmed it. They aligned the available sequences and "looked at all the ERVs known to be in the human genome." They then "checked to see if [each] site was 'empty' or 'full' in the aligned chimp sequence (R. Waterston, personal communication, April 22, 2010)." All those found to be non-orthologous were included in table 2. Again, they represent only a handful of tens of thousands. The researchers directly observed that the majority of ERVs in the human genome are in orthologous loci in the chimpanzee genome.
[The CSA consortium] also include Class I ERVs indicating the chimpanzee genome contains 234 (>1MB) while the human genome contains 5 (8 KB).
You are referring to table 2 on page 75 of
CSAC (2005). And the description reads: "Number of lineage-specific insertions (with total size of inserted sequences indicated in brackets) in the aligned parts of the genomes."
Those are only the lineage-specific, i.e. non-orthologous insertions. And how many are there in total? As I explained in my essay, and to you here, "the total length of all ~200 thousand ERVs is at least ~127 Mb (megabases; million base pairs) (
IHGS Consortium, 2001)."
But let's look at class I elements, specifically. Table 11 on page 880 shows that there are 112,000. It is painfully obvious that the 5 lineage-specific class 1 ERVs in table 2 of
CSAC (2005) represent a tiny fraction of the 112,000 total class 1 elements reported in table 11 of
IHGS (2001).
As I said, either a given ERV in the human lineage is accompanied by a corresponding ERV in the same genomic location in the chimpanzee lineage (orthologous ERV), or that given ERV is not accompanied by such an ERV (non-orthologous ERV). Thus, the remaining tens of thousands are necessarily orthologous.
And again, I corroborated this with Dr. Waterston.
The findings of the Chimpanzee Genome Consortium are consistent with Polavarapu et al who wrote their paper a year later. What these findings are not consistent with is your statement that most of the human ERVs are found in the chimpanzee genome.
I already addressed this in my essay, and here:
CSAC (2005) was looking at the entire aligned chimpanzee and human sequences, and reporting the non-orthologous ERVs, where the remainder are necessarily orthologous. And
Polavarapu et al. (2006) was looking at indels to directly isolate the non-orthologous ERVs. They are both consistent with each other, and they are also both consistent with the all the raimaning ERVs they were not directly mentioning being necessarily orthologous, since an element can only either be orthologous (in the same spot) or non-orthologous (not in the same spot).
Please read the following carefully:
My ERV Essay said:
Indel Variation Observed to Involve ERVs
Total indel variation provides a minimum number of transposable elements that can be shared in orthologous loci between chimpanzees and humans; but further examination is necessary to determine the actual number. One way this can be done is by isolating only the indels that are the right size to potentially be non-orthologous ERVs. Once this is done, the sequences corresponding to those gaps can be individually examined.
The results of such analysis are that less than 100 ERVs are human-specific (
Polavarapu, Bowen, & McDonald, 2006). As previously stated, if a sequence is not only at a given locus in one lineage, nor only at a given locus in the other, then the only remaining possible state in which it can exist is at the same locus in both lineages. So the number of orthologous ERVs is the total number minus the number that form gaps. With less than 100 of the ~200 thousand ERVs in the human genome yielding no gaps, the percentage of ERVs in orthologous loci is grater than 99.9%.
Whole-genome Analysis
In 2005, the available sequence of the Chimpanzee genome was aligned with that of the human genome, and an extensive comparison analysis was performed. As part of this analysis, the researchers looked at every available solo LTR and full-length ERV in the chimpanzee genome, and checked to see if there was also one at each corresponding locus. Just as with the examination of indels, the results were that less than 100 ERVs are human-specific and less than 300 ERVs are chimpanzee-specific (
Chimpanzee Sequencing and Analysis Consortium, 2005; R. Waterston, personal communication, April 22, 2010).
.
You made a statement that turns out to be fundamentally wrong, I would like some justification for not qualifying the statement and now dismissing the evidence.
As demonstrated above, I have not dismissed anything; I have specifically addressed what was posed to me, and the statements I made are sound.
That's not what they say, in fact, they don't mention the orthologous ERVS much at all.
Once again:
Either a given ERV in the human lineage is accompanied by a corresponding ERV in the same genomic location in the chimpanzee lineage (orthologous ERV), or that given ERV is not accompanied by such an ERV (non-orthologous ERV, i.e. specific to one or the other, i.e. lineage specific ERV).
By reporting the number of non-orthologous ERVs, they are necessarily reporting the the orthologous ones as the remainder.
What is more important is that ERVs represent something like 7% of the known divergence and they only occur in 1 live human birth out of a hundred. You are not just looking at an invasion of ERVs since the split, your are looking at major germline reconstruction and inundation. This is remarkably strange to me since we would have lived in the same geologic area (equatorial Africa) until about a million years ago, some estimates indicate a mass migration from that region about 40,000 years ago.
The percentage is as high as it is because retroviruses began all this endogenization tens of millions of years ago (
Hughes and Coffin, 2005).
Here's the problem, ERVs are nearly extinct in the human genome with only a few that actually work. The problem with homology arguments is that the necessarily allow for the inverse logic. If sequence identify being similar or identical argues for common ancestry then differences are proof against.
As explained so thoroughly in the literature I cite in my essay, that is easily accounted for by a decrease in activity. And I didn't say that sequence identity of ERVs is evidence of common ancestry. I did say that the pattern (the relation between differing levels) of LTR-LTR discontinuity (and thus, yes, sequence identity) does provide evidence. And it does so by its corroboration of the hierarchical patterns.
Transposable elements in 'Retroelements and the human genome' make up 45% of the human genome, retroelements make up 42% and LTRs including ERVs primarily make up 8%. What I would like to know is as a ratio what percentage of those have been inserted since the split, not only human but chimpanzee sequences. I am constantly fielding statements like this that don't stand up to close scrutiny. What percentage do CERV1/PtERV1 represent as compared to HERVs? More importantly, how do you determine that?
The non-orthologous ERVs are the ones that occurred after the divergence of chimpanzees and humans. As for a ratio, there are 5 lineage-specific class II ERVs in the human genome (HERV-K) (
CSAC, 2005, p.75, table 2) to a total of 112,000 HERV-K insertions in the human genome (
IHGS, 2001, p.880, table 11).
You lost me, it is specifically the commonality of human and chimpanzee ERVs that you claim as proof of common ancestry. You have made some pretty impressive statements, now I am going to need to know where the direct comparisons are.
I was under the impression that you read my essay. It is all there, with references. And why would you think I only cite ERVs shared between chimpanzees and humans, when I keep mentioning nested hierarchies? There can be no nesting (groups within groups) between only two clades. You need three or more.
Furthermore, you actually quoted the part of my essay on layer 1 of ERV evidence where I mention shared ERVs between all haplorrhines
:
When we examine the collective genome of Homo sapiens, we find that a portion of it consists of ERVs (IHGS Consortium, 2001). We also find that humans share most of them with Chimpanzees, as well as the other members of Hominidae (great apes), the members of Hylobatidae (gibbons), and even the members of Cercopitheciodae (old world monkeys) (Kurdyukov et al., 2001; Lebedev et al., 2000; Medstrand and Mager, 1998; Anderssen et al., 1997; Steinhuber et al., 1995).
In fact, every one of the three layers of ERV, as I explain them in my essay, includes patters of ERVs shared amoung all haplorrhines.
What is more the dating of the ERVs create a problem since it makes no sense.
In fact CERV 2 (Chimpanzee ERV) has an estimated age of 21.9 mya to 14.1 mya with no human orthologues. The CERV1/PtERV1 with 100 members is estimated between 5 mya and 7.8 mya. That means that the most abundant ERVs in the Chimpanzee Genome came about before or right around the split That means in addition to the adaptive evolution of the human brain our ancestors would have had to adapt their immune system to push these ERVs to the brink of extinction.
So ERVs make up 8% of the human genome, do you seriously expect me to believe that they resulted from germline invasions? If this were the case then ERVs would be the source of human diversity since they would have to be happening on a staggering scale and rapid pace.
I already addressed the "most abundant" misconception, and I also already explained how PtERV1 simply has "uncharacteristically high long terminal repeat discontinuity ratios, likely due to interelement recombination/conversion and viral transfer (
Polavarapu, Bowen, & McDonald, 2006)".
I don't know why you don't see a problem here but one thing is clear to me, this is just another homology argument that can't stand up to close scrutiny.
As demonstrated above, in the previous posts, and in my rebuttal of Jonathan M's posts, the ERV evidence for common ancestry has stood up quite well to the scrutiny it continues to undergo.
