• Starting today August 7th, 2024, in order to post in the Married Couples, Courting Couples, or Singles forums, you will not be allowed to post if you have your Marital status designated as private. Announcements will be made in the respective forums as well but please note that if yours is currently listed as Private, you will need to submit a ticket in the Support Area to have yours changed.

  • CF has always been a site that welcomes people from different backgrounds and beliefs to participate in discussion and even debate. That is the nature of its ministry. In view of recent events emotions are running very high. We need to remind people of some basic principles in debating on this site. We need to be civil when we express differences in opinion. No personal attacks. Avoid you, your statements. Don't characterize an entire political party with comparisons to Fascism or Communism or other extreme movements that committed atrocities. CF is not the place for broad brush or blanket statements about groups and political parties. Put the broad brushes and blankets away when you come to CF, better yet, put them in the incinerator. Debate had no place for them. We need to remember that people that commit acts of violence represent themselves or a small extreme faction.

Genome sequenced of extinct human relative

Blayz

Well-Known Member
Aug 1, 2007
3,367
231
60
Singapore
✟4,827.00
Faith
Atheist
Marital Status
Married
Well if you want a critique then I'd say the 1.9 fold coverage is not great. They used BWA for mapping which is good, they mapped to hg18, which is bad. Their evolutionary analytics are pretty sound, and their CIs are good, but it moves to speculation pretty quickly.

Overall I'd say if you care about your relationship to a bunch of 40000 year old almost humans, then it's an interesting paper.

It's not clear to me why anyone does care though.

sfs will have a better take, if he comments.
 
Upvote 0

Naraoia

Apprentice Biologist
Sep 30, 2007
6,682
313
On edge
Visit site
✟23,498.00
Faith
Atheist
Marital Status
Single
Well if you want a critique then I'd say the 1.9 fold coverage is not great. They used BWA for mapping which is good, they mapped to hg18, which is bad.
Well, the page at the Max Planck Institute website does quote much higher coverage. I suppose that's why the genome is newsworthy again after the 2010 paper.

It's not clear to me why anyone does care though.
Perhaps they'd feel the same way about your interests :p

sfs will have a better take, if he comments.
That would be nice.
 
Upvote 0

sfs

Senior Member
Jun 30, 2003
10,863
7,882
65
Massachusetts
✟397,974.00
Country
United States
Gender
Male
Faith
Non-Denom
Marital Status
Married
Well if you want a critique then I'd say the 1.9 fold coverage is not great.
As Naraoia points out, that was the old genome sequence. The new version has 30x coverage over the "mappable genome".

Note: my usual source(*) on matters Denisovan spends Wednesdays at another lab, so I can't quiz him on what they're doing with the improved genome. I'll just ramble instead.

(*) One of the series of sometime office-mates that I classify as "people who are smarter than I am".

They used BWA for mapping which is good,
Written by another member of the same series, I note.
they mapped to hg18, which is bad.
All analyses that made comparisons to modern humans used alignments to the chimpanzee genome, to avoid introducing population-specific biases.

Overall I'd say if you care about your relationship to a bunch of 40000 year old almost humans, then it's an interesting paper.
The current news is about the public release of a new version of the genome, not about a paper; they say they'll write a paper on it eventually.
It's not clear to me why anyone does care though.
Based on the amount of news coverage the Denisovan and Neandertal papers have gotten, many people do seem to care about our history as a species. Presumably the new genome will permit much better estimates of Denisovan diversity, of the size and distribution of Denisovan genetic variation contributed to modern humans, and of the possible role of positive selection in the preservation of some of that variation. Probably other things that I can't think (see above re relative intelligence).
 
Upvote 0

Blayz

Well-Known Member
Aug 1, 2007
3,367
231
60
Singapore
✟4,827.00
Faith
Atheist
Marital Status
Married
Written by another member of the same series, I note.

Pardon? what does "member of the same series" mean?

All analyses that made comparisons to modern humans used alignments to the chimpanzee genome, to avoid introducing population-specific biases.
Is there a reason they cannot move to hg19?

many people do seem to care about our history as a species.
I still don't get the fascination, especially not at the genetic level.

Presumably the new genome will permit much better estimates of Denisovan diversity, of the size and distribution of Denisovan genetic variation contributed to modern humans
Again, I don't get why anyone outside the research area cares.

and of the possible role of positive selection in the preservation of some of that variation.
Can you expand on this? Which phenotypic traits could be considered?
 
Upvote 0

CabVet

Question everything
Dec 7, 2011
11,738
176
Los Altos, CA
✟35,902.00
Faith
Agnostic
Marital Status
Married
Politics
US-Others
I still don't get the fascination, especially not at the genetic level.

Again, I don't get why anyone outside the research area cares.

So, you are not fascinated with the possibility of knowing how our genome changed over 40 thousand years?
 
Upvote 0

Blayz

Well-Known Member
Aug 1, 2007
3,367
231
60
Singapore
✟4,827.00
Faith
Atheist
Marital Status
Married
So, you are not fascinated with the possibility of knowing how our genome changed over 40 thousand years?

It's academically interesting sure, and if I happen to wander past a population geneticist giving a talk on the subject then I might stop and listen, even though I know I'll probably understand 2 words out of 3.

I don't understand why laity, who have no real understanding of genetics (which is not an insult, any more that it would be an insult to state that I don't understand nuclear physics) are interested. For someone with no training in the field, you'd have to dumb down the "changes over 40 000 years" to the point it is not worth it.

In addition, this research wont answer that question. Our entire knowledge of these people is a handful of tiny bone fragments. We might get some clarification on early hominid ancestry and movement, but until sfs backs up his claim that it will inform on positive selection and intelligence, it's not going to tell us anything about us.

Maybe it's because I work in applied bioinformatics. I find trying correcting disregulation in the tumor immune micro environment or determining the mechanism of infection triggered alternate splicing (as two examples) to be far more interesting.
 
Upvote 0

sfs

Senior Member
Jun 30, 2003
10,863
7,882
65
Massachusetts
✟397,974.00
Country
United States
Gender
Male
Faith
Non-Denom
Marital Status
Married
Pardon? what does "member of the same series" mean?
The series of office-mates who are smarter than I am. (Heng Li, in this case.)

Is there a reason they cannot move to hg19?
Oh, you mean the build. No clue -- I can ask.

I still don't get the fascination, especially not at the genetic level.
Lots of people have vague notions about Neanderthals, and the discovery that many of us are partly descended from them is fascinating on a pretty basic level, the level that cares about genealogy. Likewise, the discovery that there were relatives of Neandertals that that we didn't know about until this bone was found, and that we're descended from them too is also interesting.

Can you expand on this? Which phenotypic traits could be considered?
It wouldn't be looking for phenotypes, but for haplotype chunks that have increased in frequency unusually quickly.
 
Upvote 0

Blayz

Well-Known Member
Aug 1, 2007
3,367
231
60
Singapore
✟4,827.00
Faith
Atheist
Marital Status
Married
The series of office-mates who are smarter than I am. (Heng Li, in this case.)

I have a little shrine to him in my office...although to be fair he was at Sanger not Broad for BWA.

Oh, you mean the build. No clue -- I can ask.
Please do. I'd be interested in the answer. Maybe I can take their data do a UCSC liftover and republish :clap:

It wouldn't be looking for phenotypes, but for haplotype chunks that have increased in frequency unusually quickly.
Not getting that mate. There's not enough Deniosovan material to determine any kind of frequency, best you could do would be qualitative presence/absence, at which point my question would be which ethnic hapmap would you compare to? Once you have done that, how do you assign function? We can barely (or perhaps not even) do that with GWAS now.
 
Upvote 0

sfs

Senior Member
Jun 30, 2003
10,863
7,882
65
Massachusetts
✟397,974.00
Country
United States
Gender
Male
Faith
Non-Denom
Marital Status
Married
Please do. I'd be interested in the answer. Maybe I can take their data do a UCSC liftover and republish :clap:
Simple answer: it's not hg18, it's hg19.

Not getting that mate. There's not enough Deniosovan material to determine any kind of frequency, best you could do would be qualitative presence/absence, at which point my question would be which ethnic hapmap would you compare to? Once you have done that, how do you assign function? We can barely (or perhaps not even) do that with GWAS now.
I just mean the kind of analysis done in [URL="http://www.citeulike.org/group/3378/article/9707766]this paper[/URL]. The identification of loci that have been under positive selection is independent of the origin of the haplotypes, and usually doesn't depend on knowledge of phenotype, but on things like long haplotypes or skewed allele frequency distributions. What you can use the Denisova genome for is to see whether archaic haplotypes were positively selected for once they entered the modern human genome.
 
Upvote 0

CabVet

Question everything
Dec 7, 2011
11,738
176
Los Altos, CA
✟35,902.00
Faith
Agnostic
Marital Status
Married
Politics
US-Others
What you can use the Denisova genome for is to see whether archaic haplotypes were positively selected for once they entered the modern human genome.

Unless I am misunderstanding you, I don't think we can do this with just one individual. We would need a relatively large sample size to estimate haplotype frequency in the past.
 
Upvote 0

sfs

Senior Member
Jun 30, 2003
10,863
7,882
65
Massachusetts
✟397,974.00
Country
United States
Gender
Male
Faith
Non-Denom
Marital Status
Married
Unless I am misunderstanding you, I don't think we can do this with just one individual. We would need a relatively large sample size to estimate haplotype frequency in the past.
Assuming a very small number of matings produced the admixture (which seems likely), then we know that the frequency would have started out very low in the modern human population.
 
Upvote 0

CabVet

Question everything
Dec 7, 2011
11,738
176
Los Altos, CA
✟35,902.00
Faith
Agnostic
Marital Status
Married
Politics
US-Others
Assuming a very small number of matings produced the admixture (which seems likely), then we know that the frequency would have started out very low in the modern human population.

Yes, but we need a decent sample size from ancient DNA as well to rule out the possibility of incomplete lineage sorting. With only one Denisovan genome, how can we tell if the haplotype wasn't also common in the ancestor of human/denisovan?
 
Upvote 0

Notedstrangeperson

Well-Known Member
Jul 3, 2008
3,430
110
37
✟27,024.00
Gender
Female
Faith
Anglican
Marital Status
In Relationship
This study reminds me of a something I came across a little while ago:

One of the oldest skeletons in Australia is "Mungo Man" a.k.a. LM3 - originally thought to be one of the first Aboriginal Australians. He has the anatomy of a modern human and dates between 50,000-45,000 years old (although some estimates go as high as 68,000). Despite his age, a team headed by Simon Easteal managed to extract DNA from his bones and compared it to the DNA of other humans.

Surprisingly, he was not related to the famous Mitrochondrial Eve (link). They also compared him to neanderthals, ancient Aboriginals and modern Aboriginals. He wasn't related to any of them either. Some of his DNA survives in modern humans, not in their mitochondria but in chromosome 11.

There's an in-depth analysis of his DNA here:
Mitochondrial DNA sequences in ancient Australians: Implications for modern human origins.
----------------------------------------------

Long story short - is it possible this man was a Denisovan, or is something more subtle going on? We know the Mitochondrial Eve obviously wasn't the only woman alive all those years ago, so could Mungo Man belong to a lineage which simply died out?
 
Upvote 0

Naraoia

Apprentice Biologist
Sep 30, 2007
6,682
313
On edge
Visit site
✟23,498.00
Faith
Atheist
Marital Status
Single
Is there a reason they cannot move to hg19?
They did. Check the page at the MPI.
Downloads

The Denisova Genome Consortium have chosen to release the raw sequence data and alignments.



Provided here are the alignments using BWA to the human (hg19/GRCh37) and chimpanzee (panTro2) genomes.
I still don't get the fascination, especially not at the genetic level.
We get that you don't get it. Can we move on?

I don't understand why laity, who have no real understanding of genetics (which is not an insult, any more that it would be an insult to state that I don't understand nuclear physics) are interested.
Why am I interested in cosmology when I'm a biologist and I haven't done physics since I was 16? Why is anyone interested in anything? Don't know. Don't care. It just happens.

Maybe it's because I work in applied bioinformatics. I find trying correcting disregulation in the tumor immune micro environment or determining the mechanism of infection triggered alternate splicing (as two examples) to be far more interesting.
"Useful" doesn't necessarily equal "interesting" for everyone ;)
 
Upvote 0

sfs

Senior Member
Jun 30, 2003
10,863
7,882
65
Massachusetts
✟397,974.00
Country
United States
Gender
Male
Faith
Non-Denom
Marital Status
Married
Yes, but we need a decent sample size from ancient DNA as well to rule out the possibility of incomplete lineage sorting. With only one Denisovan genome, how can we tell if the haplotype wasn't also common in the ancestor of human/denisovan?
A substantial chunk of chromosome is not going to be the result of incomplete lineage sorting; it would long since have been broken down by recombination. Substantial chunks that are present at high frequency only in populations with a Denisovan component and absent in Africa are also quite unlikely to be inherited via African Homo sapiens.
 
Upvote 0

CabVet

Question everything
Dec 7, 2011
11,738
176
Los Altos, CA
✟35,902.00
Faith
Agnostic
Marital Status
Married
Politics
US-Others
A substantial chunk of chromosome is not going to be the result of incomplete lineage sorting; it would long since have been broken down by recombination. Substantial chunks that are present at high frequency only in populations with a Denisovan component and absent in Africa are also quite unlikely to be inherited via African Homo sapiens.

Ah, now we are talking, substantial chunks is where it is all at. When anybody mentions haplotypes I always think of single loci, 500bp long, but that's my population genetics background speaking I guess.
 
Upvote 0

Blayz

Well-Known Member
Aug 1, 2007
3,367
231
60
Singapore
✟4,827.00
Faith
Atheist
Marital Status
Married
Simple answer: it's not hg18, it's hg19.

Excellent.

I just mean the kind of analysis done in this paper. The identification of loci that have been under positive selection is independent of the origin of the haplotypes, and usually doesn't depend on knowledge of phenotype
Your paper disagrees with you (about phenotype). HLA in general is a special case, and B*73 is extra special in particular, a case we have known about for decade or two. It's also not fair to refer to them as "haplotypes". and it is strongly linked to phenotype, which I'd like to hope is how it made it into Science (good paper by the way, unlike the rubbish from Cell that just crossed my desk).

but on things like long haplotypes or skewed allele frequency distributions. What you can use the Denisova genome for is to see whether archaic haplotypes were positively selected for once they entered the modern human genome.
I remain skeptical (partly about whether they will find any, but mostly whether they will have enough tie in to phenotype to make a decent paper), but await the analysis.
 
Upvote 0

sfs

Senior Member
Jun 30, 2003
10,863
7,882
65
Massachusetts
✟397,974.00
Country
United States
Gender
Male
Faith
Non-Denom
Marital Status
Married
Your paper disagrees with you (about phenotype). HLA in general is a special case, and B*73 is extra special in particular, a case we have known about for decade or two. It's also not fair to refer to them as "haplotypes". and it is strongly linked to phenotype, which I'd like to hope is how it made it into Science (good paper by the way, unlike the rubbish from Cell that just crossed my desk).
I didn't mean a repetition of that exact analysis, and I wasn't referring to haplotypes because of that paper. We know of many likely instances of positive selection in the human genome, and in the majority of cases, we know about them because a particular stretch of haplotype has increased in frequency rapidly, and we know nothing about the phenotype which was involved in the selection. Eventually the phenotype can sometimes be worked out, but that effort is independent of determining whether the selected allele came from Denisovans or not.

I wasn't suggesting that this would be the most exciting research topic in the world or the most exciting thing to do with a good Denisovan genome, nor do I have any idea whether anyone is working on it. It's just an obvious kind of thing to do with the information -- a fairly minor addition to existing approaches.
 
Upvote 0