Talk:Ancient DNA

Untitled
Part of the article is from Tom Gilbert's thesis, An Assessment of the Use of Human Samples in Ancient DNA Studies, but he has given permission for it to be posted. -- Kjkolb 11:18, 23 December 2005 (UTC)

Verification
This article needs a section on verification. How do we know if the recovered DNA sequence is correct? - Connelly 09:01, 13 June 2007 (UTC)

New article: Genetics of the Ancient World - Invitation to add content
For those who edit this page on ancient DNA you are invited to add relevant journal articles and short summaries of these articles as per ancient people groups as it bears upon the Genetics of the Ancient World. Please follow the existing format. This is a reference list with short summaries that refers back to main article pages on wikipedia.

Currently, I just put up a non-academic media source (National Geographic) on the ancient Phoenicians, but will add peer-reviewed journal articles and ask you to do the same (as well as media if relevant). Please try to make the summaries readable for the non-specialist as the purpose of this article is to help non-geneticists "wake-up" to how genetics impacts their discipline, etc. Great job on the aDNA article. regards, Hkp-avniel (talk) 18:42, 31 March 2008 (UTC)

Age limit of DNA
I replaced this section:


 * Current estimates suggest that in optimal environments, i.e. environments which are very cold, such as permafrost or ice, an upper limit of around 1 million years exists.

It had no sources and I think it is utterly disproved by the work of Allentoft et al. (2012) I think this limit will rise in the coming years since Schweitzer et al. (2012) have discovered DNA and histones in T. rex and Hadrosaur bones which are probably not human contamination.Bastion Monk (talk) 20:04, 14 November 2012 (UTC)

Use of "Antediluvian".
I see the usage of the term "antediluvian" as problematic given that we're discussing a scientific topic, not a religious one, and given that this term seems confined almost entirely to a single paper published 23 years ago.

While Tomas Lindahl did use this figure of speech, a term with such obvious mythological/religious baggage seems very out of place in this type of discussion going forward and doesn't seem to have been used in any serious sense or major publication since that first and basically only usage. We have actual scientific terms as part of the Geologic time scale for various ancient ranges of time that should be preferred here.

Phreadom (talk) 09:24, 17 March 2016 (UTC)


 * I've removed the term from the header. Joshua Jonathan  - Let's talk!  04:32, 9 July 2018 (UTC)

External links modified
Hello fellow Wikipedians,

I have just added archive links to 1 one external link on Ancient DNA. Please take a moment to review my edit. If necessary, add after the link to keep me from modifying it. Alternatively, you can add to keep me off the page altogether. I made the following changes:
 * Added archive {newarchive} to http://www.454genomics.net/downloads/news-events/geneticanalysisfromancientdna.pdf

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers.—cyberbot II  Talk to my owner :Online 07:30, 21 March 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just added archive links to 1 one external link on Ancient DNA. Please take a moment to review my edit. You may add after the link to keep me from modifying it, if I keep adding bad data, but formatting bugs should be reported instead. Alternatively, you can add to keep me off the page altogether, but should be used as a last resort. I made the following changes:
 * Added archive http://web.archive.org/web/20081217110352/http://www.454genomics.net/downloads/news-events/geneticanalysisfromancientdna.pdf to http://www.454genomics.net/downloads/news-events/geneticanalysisfromancientdna.pdf

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers.—cyberbot II  Talk to my owner :Online 03:45, 30 March 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 3 one external links on Ancient DNA. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20120331032245/http://csk.umed.lodz.pl/~dmb/DNA8/doc/?plik=inv to http://csk.umed.lodz.pl/~dmb/DNA8/doc/?plik=inv
 * Added archive https://web.archive.org/web/20110202094544/http://adna2010.com/ to http://www.adna2010.com
 * Added tag to http://www.unifi.it/dbalan/CMpro-v-p-166.html
 * Added archive https://web.archive.org/web/20091214080708/http://www.egyptologyonline.com/using_dna.htm to http://www.egyptologyonline.com/using_dna.htm

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers.— InternetArchiveBot  (Report bug) 15:01, 12 October 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified a link #14 on Ancient DNA. Please take a moment to review. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2556161/

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers. --Arina.clavel (talk) 03:10, 3 November 2020 (UTC)

90s claims
The section on 1990s discoveries needs a complete re-write. It's made up nearly entirely of "claims" of aDNA of millions of years and none of these have been validated. The section reads more like a Jurassic Park episode than a science article.--Tallard (talk) 15:53, 19 September 2018 (UTC)
 * agree. I had to keep re-reading it to make sure it wasn't me with sand in the gearbox. JohndanR (talk) 02:34, 19 October 2019 (UTC)
 * Yep. It needs a reality check. Rowan Forest (talk) 04:30, 19 October 2019 (UTC)

Evidence of proteins, chromosomes and chemical markers of DNA in exceptionally preserved dinosaur cartilage
Could you please integrate the findings of this new study, reported on here: "Cartilage cells, chromosomes and DNA preserved in 75 million-year-old baby duck-billed dinosaur" and add a short entry for it to the January section of 2020 in science? The report says:

"These bind specifically to DNA fragments in extant material, and some of the isolated dinosaur cells showed internal, positive binding in the same pattern as seen in modern cells, suggesting some original dinosaur DNA is preserved [...] The possibility that DNA can survive for tens of millions of years is not currently recognized by the scientific community. Rather, based upon kinetic experiments and modelling, it is generally accepted that DNA persists less than 1 million years. These new data support other results that suggest DNA in some form can persist in Mesozoic tissues, and lay the foundation for future efforts to recover and sequence DNA from other very ancient fossils in laboratories worldwide."

There should probably be a new section under "History of ancient DNA studies" for 2020s.

And the information should probably also be added to the article Hypacrosaurus.

--Prototyperspective (talk) 21:00, 24 March 2020 (UTC)

Terms Stone Age and Iron Age need clarification
In the last sentence of the Results section of this article, the terms Stone Age and Iron Age are used to refer to periods including 2000 years ago and 300-500 years ago respectively, following the cited reference. These dates seem to disagree with the dates for the stone ages referred to in African archaeology or those for iron metallurgy in Iron metallurgy in Africa. Perhaps the offending terms should be removed, or clarified?

Degradation kinetics
The paper by Allenloft et al nicely shows the degradation kinetics for mtDNA from Moa bones. However, the starting size appears to be 15,267 bp, for which they show that the average length at 10K years is 683bp, and at 6,830,000 years is 1bp. Nuclear genomes are order of magnitudes larger than mitochondrial genomes (3,209,286,105 bp vs 16,569bp for the human genome), so those kinetics should be recomputed from and for nuclear DNA as it is likely that the average fragment size at 6,830,000 years will be far higher than 1bp, even with a degradation rate twice as great. Pcauchy (talk) 00:57, 8 July 2020 (UTC)
 * If I understand what you are suggesting, no, that is not how it works. It is probabilistic, not absolute. It is not a situation where you get one 'degradation event' in each piece of DNA in a set time period, no matter how long it is.  After a given period of time, you will have one event every 10,000 bases and your average size will be 10,000.  In twice that much time you will have two events per 10kbp, giving a 5kbp average, etc.  The average fragment length after any given time will be the same, independent of the starting DNA length. Agricolae (talk) 01:30, 8 July 2020 (UTC)
 * I understand what you mean. So essentially like a nuclease that cuts every 10kb in your first example. However, you also hint that it is a bit more complicated than this and it probably isn't that you then get two events per 10kb, that is why the shorter fragment half-lives are higher as shown in the paper. If the degradation rate is per bp and constant (which in all likelihood it isn't), then your resulting fragments will be less prone to degradation, thus higher half-lives, that is why it is not linear in the paper if I understand it correctly.


 * But while on average, after the first "round" where on average there will be a degradation event every 10kb even with nuclear DNA, in practice the fragment size will have a very high variance (a bit like with e.g. DNaseI where you will get an average fragment size however with low digestion you will still get high molecular weight fragments when you run it on a gel, and it will be a smear).


 * The fragment size variance will be far higher for nuclear DNA than mtDNA, and in all likelihood this will also mean higher average fragment size. So on a gel, the smear will be far larger for nuclear DNA. That is what I meant. In practice, it takes far longer to digest nuclear DNA than mtDNA with the same amount of e.g. DNaseI precisely because of this. You see it with e.g. whole-cell DNase-Seq where the coverage is much higher on chrM than on the rest of the genome (and not only because there are more copies, but because there are comparatively more of the shorter fragments which are the ones being sequenced, not the long ones, and if you look at the average fragment length on chrM it is far lower at least in my experience) precisely because of this. It will take far longer to get down to an average of 1bp with nuclear DNA than mtDNA when digesting with DNaseI (which one could really compare to this process) because of the higher starting size. Yet yes this is partly because the amounts of DNaseI are then comparatively greater, however even with comparatively proportional amounts of DNaseI, it would most probably take longer to get down to 10kb average in the first place with nuclear DNA precisely because the starting size is higher and that the resulting average fragment size and variance will be greater from the start even though the cut rate will be the same, e.g. every 10kb, if this makes sense. What is for sure is that the smear will be larger for nuclear DNA than mtDNA because the starting size is far greater, and this is my entire point.


 * That is also why one needs to sonicate the chromatin longer for mammalian cells than Drosophila to get down to the same average fragent size precisely because the genome is much larger, and this is using the same sonication conditions (apart from cycle number)


 * So the question is, when starting from much larger genome sizes, does the model also account for this? That okay, there's a degradation of 5.5x10-6 per bp, but that a higher starting size will mean, in probabilistic and practical terms, most probably mean far higher fragment size variance for nuclear DNA? And it is those higher sized fragments that may stand a better starting chance to stand the test of time? How many would one need? How much starting material would one need to get 1 after x years? From what I understand it does not. At the very least it could be recomputed from fossil nuclear DNA samples to obtain the correct degradation rate which is still unknown for nuclear DNA, and to determine whether this is happening or not, i.e. whether the fragment size variance (and possibly average) is higher or not at k years. Pcauchy (talk) 14:06, 8 July 2020 (UTC)


 * This mixing apples and oranges on several levels (e.g. there is a substantial component of long-term degradation that is chemical, not enzymatic, and neither is analogous to sonication where length-dependent viscosity plays a role), while degraded genomic DNA having a wider potential size distribution doesn't affect the average size and this higher variability tends to regress toward the mean the longer the degradation progresses. Anyhow, this is a Wikipedia article, not a doctoral dissertation. The short answer is that the decay rate for mtDNA is comparable as a first-level approximation for that of genomic DNA, close enough for the purposes of a non-specialist general reference work such as Wikipedia, and it is not our job to recalculate anything; indeed, we are prohibited from incorporating such original research into articles. Agricolae (talk) 15:08, 8 July 2020 (UTC)


 * I might as well have quoted e.g. copper phenantroline which will produce similar results on naked DNA as nucleases, and is chemical-, not enzymatic based, so I don't think illustrating a point using a practical example is mixing apples and oranges at all. There are other chemical agents that will cleave DNA similarly (e.g. CuII:thiol, Ag+ etc), let alone UV radiation etc. I am not including any original research or even review material in here which we all know we are prohibited from including, but merely speculating on the consequences of higher size variability of genomic DNA which you also seem to agree on, as a result on the long term degradation kinetics, as to which extent and after how long it regresses towards the mean is probably dependent on the starting size and precisely the point of discussion. In other words, if some longer fragments can potentially exist for longer than for mtDNA. I am not asking anyone to recalculate anything in Wikipedia, but rather bringing it up for the scientific community, although obviously a better place for this would be in a comment in a peer-reviewed journal, but it doesn't pay to write it here as far as I know. If anything, one probably should not only write that nuclear DNA degrades twice as fast as mtDNA (and verify that it is indeed nuclear DNA and not naked genomic DNA) and one might as well write that a main difference with mtDNA is that the latter is naked, unlike chromatin, and is also more A/T rich, which is less rigid more fragile than G/C rich DNA, for example. Pcauchy (talk) 16:39, 8 July 2020 (UTC)


 * Besides, sonication is probably a good analogy (even though starting material, cavitation volume and surface as well as probably G/C content have a far higher impact than length-dependent viscosity) as e.g. your chr1 and chr22 fragments have the same median size, however the standard deviation is greater for chr1. So based on size, it also regresses differentially towards the mean (e.g the sharper the smear), and it might be a similar process with DNA degradation, that is all that I am trying to bring up. If anyone is willing to undertake the task to introduce a size bias parameter in a such a model using nuDNA and get it published, it might make for a more suitable study to reference than the current one. Pcauchy (talk) 17:00, 8 July 2020 (UTC)


 * I am just thinking about that grant application - 'I am requesting money for this study because the Wikipedia article on aDNA needs a better reference'. Agricolae (talk) 18:14, 8 July 2020 (UTC)


 * Let's see what Bailleul et al turn up when they try to sequence what they have found, who knows, "that grant" might just be far easier to get then for whoever goes for it. I must say I agree with Prototyperspective in that the article might need a new section mentioning those findings. It's not as if those samples could easily have been contaminated with chicken collagen. That said, they're not showing how anti-chick collagen reacts with e.g. mammalian collagen as a control. What is a bit strange too is why would one of the "cells" that they are observing be stuck in the middle of mitosis; unless the poor thing got incinerated and turned to glass or freeze-dried where it stood why would an incomplete cell division not complete and be stuck in eternity? The good news about their "DNA" findings is that propidium iodide binds only double-stranded DNA (and base-pair stackings are more stable than single base pairs), that DAPI binds not just to single base-pairs but regions of AT-rich DNA, so it might be that whatever this is is greater than single base-pairs. The bad news though is that obviously one needs at least 16 for unique hits in a human-sized genome, ideally more (e.g. 20-30) to rule out contamination, which it could still be (as in human chrY in the case of Woodward et al), there is only one way to know. Now seriously this is addressed to Allenloft et al. who cynically emphasised the "extreme improbability" of finding DNA from cretaceous dinosaur bone fragments, and I am surprised that no reviewer picked on that based simply on mtDNA, and am surprised that no one wrote anything on what I mentioned as a comment anywhere about this, after all it was easy enough to debunk e.g. Woodward et al with a single comment in Science Pcauchy (talk) 21:24, 8 July 2020 (UTC)

Table for degradation of DNA, ancient species and what not
The article as-is is quite ok, in my opinion. Could be more details but right now it is ok.

What I would love to see is a table as an overview, though. Something that you could quickly point at, and that may include things such as a representative table of old genomes, such as dinosaurs, and other extinct species. Ok, dinosaurs may be a bit difficult, but what about the wooly mammoth? How much of the DNA has been deduced correctly? I think a table would be useful, if only to quickly see the limits that we may have. And who knows, perhaps some DNA may have given an imprint in ancient mosquitos, so perhaps some stretches of dinosaur DNA may be known eventually. 2A02:8388:1602:6D80:3AD5:47FF:FE18:CC7F (talk) 18:23, 5 May 2021 (UTC)

Paper that shows no significant correlation between age and DNA fragmentation with large numbers of samples
The kinetics of DNA degradation are not very well understood to date. In addition to previous comments to moderate statements on DNA degradation kinetics, there is a clear source that could be added that shows no significant correlation between sample age and DNA fragmentation, using a collection of 185 samples Pcauchy (talk) 15:38, 6 October 2021 (UTC)

How different is hDNA from aDNA ?
suggests that hDNA, or "DNA obtained from museum voucher specimens (almost always collected during the last 200 years) has emerged as a separate subdiscipline, distinct from the study of aDNA." Should this be discussed? MaryMO (AR) (talk) 14:55, 28 June 2022 (UTC)