- Options
- Login
- Desktop Style
Options
Biocodicology - A Deeper Dive
- Biocodicology - A Deeper Dive
-
RE: Switching over to protein
MichelleL11 > 31-12-2021, 09:12 PM
Here is the full citation of the next paper:
Noninvasive species identification of parchment
Sarah Fiddyment, Bruce Holsinger, Chiara Ruzzier, Alexander Devine, Annelise Binois, Umberto Albarella, Roman Fischer, Emma Nichols, Antoinette Curtis, Edward Cheese, Matthew D. Teasdale, Caroline Checkley-Scott, Stephen J. Milner, Kathryn M. Rudy, Eric J. Johnson, Jiří Vnouček, Mary Garrison, Simon McGrory, Daniel G. Bradley, Matthew J. Collins
Proceedings of the National Academy of Sciences Dec 2015, 112 (49) 15066-15071; DOI:10.1073/pnas.1512264112
The paper is available You are not allowed to view links. Register or Login to view.
Note that this a switch from measuring DNA sequence to measuring protein sequence.
1. What is the research question that Fiddyment et al. is asking?
Can non-invasive protein extraction from a manuscript provide sufficiently intact collagen peptide samples to be able to distinguish what species was used?
A quick word on terminology – fragments of a protein can be called “peptides.” Generally, a full length protein can be called a “polypeptide.” Because there is no set threshold amino acid number for when a peptide becomes a polypeptide, you have to look in the publication to see how the authors are using the terms.
Comparative highlights of this publication:
(i) What is being tested? 13th century manuscripts rumored to have been made from “uterine” vellum (very thin/fine -- theorized to be made from fetal skins of sheep or cows – but likely unsustainable so theories such as use of rabbit have had been presented) – significant volume 220 folios from 72 different manuscripts
(ii) How are the samples prepared? Rubbing of PVC eraser to lift peptides from the manuscript surface by static charge (e.g. eraser crumbs, noninvasive)
(iii) The use of ZooMS peptide mass fingerprinting. ZooMS stands for zooarcheology by mass spectrometry.
What is mass spectrometry (e.g. peptide mass fingerprinting)? A technique that has been used in archeological collagen peptide analysis at least since 2009. Buckley et al. (You are not allowed to view links. Register or Login to view. only available for free) had previously identified 92 peptide markers in collagen that allow for distinguishing 32 species from each other.
At a very high level, protein is extracted from the eraser crumbs; the polypeptides are turned into peptides using trypsin (an enzyme that clips proteins at certain common amino acid sequences); then peptides are analyzed as to their mass to charge ratio (e.g., mass spectography) that results in unique graphical “fingerprints” when the various peptide mass/charge results are graphed. The appearance of the graphs depends on the peptide fragments produced and the specific amino acids present in the peptides.
In this example, the collagen fingerprint is changed because of a single difference in an amino acid between cow and goat (marked in black). This shifts the second peak over in the “fingerprint” allowing goat collagen to be distinguished from cow collagen. If you want more details about mass spectrometry and collagen, see this paper by Collins et al. You are not allowed to view links. Register or Login to view. (which is where this graphic came from). One thing I don't like about this graphic is that it misrepresents the reason for the "height" of the fingerprint graph peaks -- that is due to the amount of the peptide component found, not its length (which is sort of related, since longer peptide = more components) -- but it is a cartoon and useful for a beginning understanding.
You are not allowed to view links. Register or Login to view. also, is a very detailed review from the Broad Institute for those that are really, really interested.
2. What were Fiddyment et al.’s results?
Here is an example graph from Fiddyment et al. – just so you can see what these look like. The graphs show mass/charge ratio across x axis and relative abundance on the y axis.
The results indicated that 68% were calf, 26% were goat, and 6% were sheep. They found no evidence for rabbit or squirrel or anything unusual. They also found no evidence of specific proteins that would be found only in modern uterine (fetal) calf skin. See discussion in the supporting information You are not allowed to view links. Register or Login to view.. Thus, it does not look like fetal animals were used in the production of these manuscripts and the whole thing was myth (or marketing, LOL).
Remember this technique works most easily when you know the sample is relatively pure (e.g. almost entirely collagen). This is because you’ve chopped the proteins up and lost the general reference of the full sequence in the prep. This can degrade into a state of diminishing returns when you’re working with complex mixtures of proteins.
Note also that there is no amplification in this process – so if you’ve got tiny amounts of protein you’ve got a bigger problem than with DNA or RNA sequencing (where the samples can be amplified using PCR).
TLDR: ZooMS can work for distinguishing species from each other using a noninvasive manuscript sampling technique. And “uterine vellum” is just specially processed vellum that can derive from calf, goat, and sheep and NOT fetal calves.
There are general issues of practicality with this technique even as applied to collagen, but these have been addressed relatively recently using software analysis approaches see, Hickenbotham et al. (2020), You are not allowed to view links. Register or Login to view.. MALDI-ToF is just a kind of mass spectrometry. So you can draw a general conclusion that probability-based software has come to the rescue in analyzing these large data producing sequencing results so it is like next generation sequencing (NGS) in this regard.
Final note: this analysis has been performed on the Voynich Manuscript in 2014 and 10 folio parchments were identified as coming from cow, while the result from the cover was inconclusive. Later, using non-biocodicological methods, the cover was identified by Jesse Meyer of Pergamena (a parchment producer) as being goat. These results haven’t been published in the scientific literature but are generally described in: P., E. Mysak, J. Stenger, M.-F. Lemay, A. Bezur and D. Driscoll: Physical Findings in: Clemens, Raymond (ed), with introduction by Deborah Harkness: The Voynich Manuscript, Yale University Press, New Haven and London, 2016., pp. 23-37 and discussed on Rene Zandbergen's You are not allowed to view links. Register or Login to view..er I -
RE: Biocodicology - A Deeper Dive
R. Sale > 31-12-2021, 11:27 PM
Still seems like too much of a stretch to me. Parchment is one thing. Anyone can find the parchment. Finding human DNA is another matter. Then being able to determine who that DNA belonged to, that is just orders of magnitude of increased difficulty, it would seem. -
More Non-invasive Extraction
MichelleL11 > 01-01-2022, 08:05 PM
We wouldn’t be fully covering what’s out there if we didn’t discuss the work of Zilberstein et al. Also, there has been some interest in detecting “author-related” information for the Voynich, and this lab’s work does have a focus on these kinds of questions.
With a dash of “mad scientist,” Zilberstein and his lab, in collaboration with Righetti, have developed an alternative method to the PVC eraser approach for mining manuscripts for protein. You are not allowed to view links. Register or Login to view. is a lay article from 2018 that provides some interesting backstory to the development of this method (plus a bit of some other work).
We’re going to take a closer look at one of these papers I can get a full copy of (many are behind paywalls):
Maestro, Marguerite, and Morphine: The last years of the life of Mikhail Bulgakov; Gleb Zilberstein, Uriel Maor; Emmanuil Baskin, Pier Giorgio Righetti; Journal of Proteomics; 131: 199-204 (2016).
Copy of paper is You are not allowed to view links. Register or Login to view.. I did have to take a short survey and decline a “premium” account to get the PDF. I believe I previously signed up for an account on this site, but I don’t remember.
Also, note this is from approximately the same time period as the last two reviewed papers.
1. What is the research question Zilberstein et al. is asking?
Can a chromatographic bead-based extraction process allow for analysis of a relatively recent document for chemical (or protein) traces of its author’s health issues?
Comparative highlights of this publication:
(i) What is being tested? A handwritten manuscript from 1936-1940 (10 pages pulled randomly from 127 available) by Mikhail Bulgakov.
But note that other much, much older documents have been successfully analyzed using these processes but a true understanding of the impact of advanced “age” of the sample on this extraction method is still a question to me – if I could get a full copy of this You are not allowed to view links. Register or Login to view. article (by Righetti et al.), some discussion of that issue might be in there. This is of particular concern since chemical/protein based analysis does not have amplification available.
(ii) Use of ethylene vinyl acetate sheet (ironically, “EVA”) impregnated with anion and cation exchangers (SCX and SAX, strong cation or anion exchange, respectively) and hydrophobic resins (combination pulls ANY KIND of charged molecules to it) to get the chemical/protein sample. Non-invasive. Combination is laid upon the manuscript folio and then pulled off. See cartoon below from summary of prior referenced Righetti et al. review.
It seems the methods illustrated above is the more mature version of the extraction method (circa 2020) – it appears that for this paper in 2016, only a small aliquot of a combination of SCX beads and crushed hydrophobic resin (C8) was directly placed on the manuscript (see Figure 1 of paper).
(iii) Use of GS-MS (Gas-liquid chromatography mass spectrometry). This is still mass spectrometry, but the initial sample preparation differs from what was used by Fiddyment et al. (e.g. MALDI-ToF vs. GS-MS).
2. What were Zilberstein et al.’s results?
By comparing the chemical fingerprints to known samples of morphine from that time, that may have degraded similarly (found in police and pharmacy storage) and modern morphine – they were able to match and identify morphine and its metabolites on each page. See Figure below.
They also state they found biomarkers for the degenerative disease that Bulgakov is believed to have been suffering from -- “nephrotic syndrome,” a kidney disorder (data not shown). This last part is pretty speculative – and is presented like there is going to be a further paper about it, but I couldn’t find any additional publications.
TLDR: Zilberstein et al. found traces of morphine and its metabolites left on the pages of a manuscript, supporting a possible conclusion that the author Bulgakov was taking morphine at the time these pages were produced.
Although the use of this morphine at this end of life stage for the author is hypothesized to be due to a possible nephrotic syndrome, it is also know that the author had been addicted to morphine earlier in life and even wrote a book about the experience. Thus, it is possible that these traces are evidence of a continuation (or relapse back into) the morphine addition that had been presented as having been “left behind.”
I note that Zilberstein is known to be a huge fan of Bulgakov's work and the whole (rather obvious in my view) idea that the morphine use could be a relapse rather than something due to pain relief from nephrotic syndrome, or admittedly maybe both, was not even mentioned. This is an example of the looser scientific writing that is mentioned below and feels like a slippery slope to me into possible "theory-casting."
How this kind of research would apply to the Voynich is obviously more speculative than other analyses we have discussed. There is the ever present issue of attribution – traces of many molecules could be found (DNA, RNA, protein and as in this publication, lots of other chemicals) and whether these were due to an exposure that occurs at the time of the manuscript’s production or anytime during the 600 year existence of the manuscript would involve a lot of speculation. But this does provide another non-invasive extraction method that could be useful in providing chemical/protein evidence in trying to pinpoint the origin and/or characteristics of the scribes (perhaps author(s)) of the Voynich.
Finally, I also find the work and reporting done by these two labs to be a less “standard” scientific approach than the prior reviewed papers – which could be interesting but seems to me to be a bit more of a risk to the manuscript – but that could just be my own personal biases talking. -
RE: Biocodicology - A Deeper Dive
Mark Knowles > 02-01-2022, 11:51 AM
I suppose another issue is what quantity of DNA would the scribe have deposited. Some of this DNA will have degraded to such an extent that it is no longer useful. I am inclined to the view that if there are sufficient quantities of scribal DNA surviving then it would be possible to extract a sample without damaging the manuscript. However I would think the scientific techniques and know-how will have to develop further before this becomes possible. -
RE: Biocodicology - A Deeper Dive
Mark Knowles > 05-01-2022, 01:05 PM
An important question I would think is what advances are needed in DNA science in order to extract the scribe(s)' DNA without harming the manuscript and what advances are needed in DNA science and increased public population DNA databases to identify the scribe(s) and their characteristics. And how long will it be before these advances become reality? (Are there any advances required in sequencing techniques and potential isolating specific DNA when many samples from different people are present?) If DNA is found from more than one scribe it could be compared to see if they are related or of the same gender. -
RE: Biocodicology - A Deeper Dive
Mark Knowles > 05-01-2022, 01:35 PM
It worth noting that whenever I look into different aspects of this subject it is clear, maybe unsurprisingly, that the science is developing fast. Many things that could be seen as pure fantasy 20 years ago are reality now, so it is not unreasonable to consider what may be reality in 20 more years.
One thing I think it is important to note is the idea popularised in the book and movie "Jurassic Park" of extracting DNA from dinosaurs so as to bring them back to life now appears to be impossible, because in 60 million years the DNA would be so degraded as to make sequencing impossible. So there are genuine limits as to what is possible and what is not. However the mere 600 years, by comparison, of the Voynich is clearly a hugely shorter timeframe, though touch DNA maybe smaller in quantity and may degrade much more rapidly than DNA extracted from bone, though I see no reason as yet to believe this DNA will be too degraded.
I am generally even more optimistic about what can be done once the DNA is sequenced. Once that is done we move from the very specific problem pertaining to extracting and locating specific touch DNA from an ancient manuscript to a very general problem of identifying an individual from their DNA on the basis of how they are related to other individuals from a very large human database. I would have thought these techniques will continue to develop rapidly. Likewise determining characteristics from DNA is an area which will continue to develop rapidly. -
RE: Biocodicology - A Deeper Dive
Bernd > 06-01-2022, 10:09 PM
DNA is extremely stable if stored under the right conditions. Likewise it degrades and fragments fast under unfavorable ones. And the surface of parchment isn't exactly the best place. Surely not the worst but far from ideal.
There's no doubt it would be possible to extract DNA fragments of many people who have handled the manuscript over time. However distinguishing between these people and building a DNA profile of the scribe(s), most likely the oldest and most degraded human genetic material present in the VM, seems a bit of a stretch to me.
New sequencing methods bring increasingly powerful techniques to extract and sort DNA of countless organisms from environmental samples but I am unsure how far this could be deployed on minute traces of highly fragmented DNA of multiple humans. If there was a single individual creating and handling all the bifolios over a long time maybe. If we are indeed dealing with a collective of scribes and maybe illustrators each creating a part of the VM within a short time things get really hard.
But theoretically these 2 hypotheses could be tested by taking samples from all bifolios and comparing DNA profiles. If there are significant differences between bifolios attributed to different scribes this would be an indication for a multi-scribe scenario. If we see one massive DNA profile across all bifolios this would rather speak for a single author. All providing later contamination could be removed. This however likely would require a destructive approach.
First and foremost anyone interested (and capable) of carrying out such analyses on the VM must demonstrate the feasibility and robustness of the proposed methods and the significance of potential results in relation to the expected damage to the manuscript. The boundaries of current technology could relatively easily be tested on less valuable documents, same for the proteome of the parchment or albumen in paint. I'm sure a lot is possible and even more will be in the near future but the VM is not the right proving ground for experiments.