Homology studies. A DNA strand 17 bp long is pretty distinct. If a strand that long or longer was found, we would gain a lot of information about the evolutionary history of dinosaurs that we could never get from fossils.
However, the article isn't clear about what they found. It says DNA, then it says there may not be a recoverable sequence. In that case nucleic acids would be a better term than DNA. Without more info on what stain they used to identify the "DNA", the implications of this research are hard to figure out.
What they found is structures that looked like cells under a microscope: https://academic.oup.com/view-large/figure/200008005/nwz206f... Section C shows cell-like structures and D shows the hypothesised DNA-like structure. They compare with a modern Emu's microstructures in section G.
There was a paper not long ago saying that the DNA half life is about 500k years or so and that most of the DNA would be degraded.
The issue I had with this is why can't we reassemble chunks of DNA like we can with a puzzle.
I'm certain specialized algorithms exist within DNA analysis for this reason. This isn't my field and there are already algorithms that exist on the top of my head that could solve this issue.
The main issue is that you'd have to find a LOT of DNA to build one complete strand and you wouldn't have one strand from one specific animal.
That's the halflife of each bond between nucleotides. After one tenth of a half life you're left with strands of average length ~20 nucleotides. After a full half life they are of length ~2, i.e., mostly just scattered GCs, ATs, etc. After several half-lives, it's almostly completely single nucleotides, and the chance that there are any surviving strands longer than a given length N falls exponentially in time and N.
What does half life of 500,000 years mean? Does that mean any given gene will, on average, split every 500,000 years? Or become unusable in 500,000 years?
65 million years would mean 130 halvings. Basically nothing left, then.
...but this may depend on environmental conditions. this must depend on average environmental temperature, right? Might dinosaur specimens in Antarctic or Arctic regions have experienced significantly lower average environmental temperatures? That could increase the half life significantly.
Maybe we'll find Cryolophosaurus DNA some day...
EDIT: the DNA degradation study specimen was in the south island of New Zealand, which had an average burial temperature of 13C.
Antarctica along the coast has an average annual temperature of about -10C. Further inland and at higher elevation, it's much colder and dryer. Both those increase DNA stability. It's not crazy to me that you might have a factor of 100x greater stability for specimens found where the average temperature over the last 65 million years was 20-40 degrees colder and also dryer.
EDIT AGAIN: Just to support what I said, it seems like a difference of average burial temperature of just 2.5 degrees C doubles the half-life of DNA (and this is multiplicative), so with a 20 degree difference in average burial temperature, that would be 8 doublings of half-life length (factor of 256), so instead of 500,000 years until the DNA was all broken up, you could have 128 million years. https://figshare.com/articles/_Predicted_DNA_half_life_for_v...
so indeed a difference of maybe 40C should lead to a difference in half life of about 20-30%. I think about 30 half lives (~10^-9 degradation factor) should be the limit of recoverability, so perhaps 521x30 = 15k years in normal conditions up to 521x30x1.25 = 20k years in low temperature.
Of course, other factors such as humidity could contribute as well to the half-life, and reactivity is a lot more complicated than the Arrhenius model in reality. But even then I would be surprised such a vast difference in reactivity outside of true cryogenic conditions (maybe even shielding from radiation?).
edit: your second edit is interesting :) is that curve polynomial or exponential?
Oh yes, I definitely messed up my calculations. Half life is (inversely) proportional to the reaction rate itself I guess (I were associating it with the exponent, because half life is usually in the exponent), so the half life itself would be proportional to exponential reciprocal of temperature (e^(a/T)).
This means significant differences in HL for minor temperature variations under e.g. Arrhenius model (consistent with your graph I think). To extrapolate it some parameters need to be estimated though, which sounds interesting, I'll get around to that later...
You raise a very good question. Unlike C14 dating, where the halflife is independent of environmental conditions, DNA is a chemical molecule, whose decomposition is a chemical reaction, dependent on exposure to time, temperature, pressure, light, and chemistry of the environment.
I was hoping the article would explain why the conditions would be so favorable to make it possible.
Right. There must have been trillions of dinosaurs over the eons. Perhaps some individuals were buried in areas with colder conditions and chemical preserving agents. It may take a long time for us to find out.
That's pretty much how modern genome assemblies work. We currently do not have a reliable or accurate way of sequencing DNA longer than ~200 bases. So we fragment the DNA into small pieces, sequence them and then put it all back together.
It's not a completely solved problem as there are features of genomes that can make this process difficult or complicated (repetitive regions, highly heterozygous organisms, etc). Especially with short sequences.
We have technology right now that can allow us to sequence long fragments, but at lower quality and accuracy. There are a lot of tools out there that uses the longer, but lower quality sequences to scaffold the shorter, higher quality sequencing data.
That’s a good enough analogy for modern DNA sequencing.
The code of DNA is an image trapped in an opaque box. Instead of being able to image the complete picture of an organism’s DNA, you take thousands of identical images in opaque boxes, smash the boxes and their contents into tiny pieces to get bits of image out, then line up the now exposed fragments to see the original image.
It is the realignment that is the computationally expensive part of the process.
I mean that a soup of loose nucleic acids can't be sequenced even if we wanted to, we can't learn anything about their genome unless there's actual intact lengths of DNA.
That quote implied the former to me, so- there's not much to be learned from just nucleic acid soup
Even this amount of genetic material is a fantastic find, though. Maybe there's hope of some future technique being able to recover short sequences somehow.
However, the article isn't clear about what they found. It says DNA, then it says there may not be a recoverable sequence. In that case nucleic acids would be a better term than DNA. Without more info on what stain they used to identify the "DNA", the implications of this research are hard to figure out.