The Voynich Ninja

If someone knows of a spot or spots in this very long manuscript where it looks like the scribe may have left a small hair or a drop of saliva or a drop of blood or a fleck of skin or a sliver of finger nail or a teardrop or mucus or pus or some other kind of human biological material that would be quite significant. If there is such a point or points where the scribe deposited a large amount of DNA on a page of the manuscript finding such a point with so very many pages of the manuscript would be hard.

The next set of materials in this "deeper dive" is a bit of a smorgasbord of media on this topic.

First, I'd like to direct everyone's attention to the You are not allowed to view links. Register or Login to view. project which has a stated goal "to document the biological and craft records in parchment in order to reveal the entangled histories of improvement and parchment production in Europe from 500-1900 AD."  It obviously includes a bunch of people whose research has been reviewed in this string and work that comes out of this project could be helpful in the ultimate completing of the biocodicological analysis of the VM.  I don't know how long the funding for this project is going to exist, but it is nice to see a cross-functional group working on these issues.

Next is a podcast episode in You are not allowed to view links. Register or Login to view., specifically You are not allowed to view links. Register or Login to view., entitled Biocodicology: From Dust to Data (December 2021).  The two guests are authors of papers we have reviewed or know about, Sarah Fiddyment (see, multiple reviewed papers) and Timothy Stinson (the author of the 2009 paper that started the board's look into biocodicology), and it is interesting to hear them provide general overviews and discuss specific details that don't necessarily get into the publications.

While you're at Coding Codices, you can also listen to board member Lisa Fagan Davis' interview in their You are not allowed to view links. Register or Login to view. -- although I've obviously read Lisa's You are not allowed to view links. Register or Login to view. concerning the differentiation of the various scribes in the VM a number of times [it's not strictly biocodicology but really interesting], I enjoyed listening to stories from work involved in collecting data and putting together that publication so I'm going to recommend it here.

Finally, if you have an additional 30 minutes to spend on this topic, here is a You are not allowed to view links. Register or Login to view. to a short talk by Sarah Fiddyment.  

Each of these presentations is not to the depth that our paper reviews have done, but I do think it makes sense to look at overview materials to help solidify understanding and also get to know a bit better some of the researchers that are involved in this work.  If any questions come up after look at these presentations -- feel free to ask.

I'll be back when I can with a few more summaries of very recent papers to wrap up the board's "deeper dive" into biocodicology.

The reading for today discusses the use of the "next" next generation sequencing (NGS) -- sometimes this is called "third" generation and sometimes its called "fourth" generation sequencing -- depending on how many "technology generations" the authors recognize.

The technique is nanopore sequencing, and a very high level summary of how it compares to "Illumina" sequencing (generally categorized as a type of NGS) is available You are not allowed to view links. Register or Login to view..

I haven't posted a whole series of articles in this sub-area, because in my view I think this has less of a chance of being applicable to the Voynich Manuscript (VM) -- but I could be wrong (see speculation below).  Plus, knowing about all the technologies available and their pros and cons is helpful.

A good overview to this technique as it applies to historical document analysis is You are not allowed to view links. Register or Login to view. by Sterflinger and Piñar (2021).

They have used nanopore technology in a series of publications that are summarized in this book chapter.

Importantly, the focus is on building data for all DNA present in the manuscripts/artifacts.  

This means not only source species for parchment and human DNA from handling but DNA from all the organisms/microorganisms present -- which are a lot.  This has immediate implications for conservation and the Piñar lab made its name in 2015 You are not allowed to view links. Register or Login to view. of a common conservation issue -- microorganism communities that result in purple spots on the pages -- and this lab is currently using nanopore technology on these kinds of analyses.

At a high level, nanopore sequencing technology involves running the unwound DNA sequence fragments through a very, very small hole and identifying the bases by the change in charge.

[attachment=6183]

Because this technique does not involve amplification but just sequences what is available, it does remove "bias" that is induced by the polymerase chain reaction (PCR) aspects of particularly Sanger (first gen) sequencing, but also NGS -- which involves some PCR amplification during the library formation.

The lack of PCR does mean that sequence integrity is more important, so for old DNA, that can be an issue.  However, software can help to "bridge the gap" between both (1) degraded DNA and (2) the inherently greater amount of errors in the nanopore system.  Of course, all software needs reference sequences to be useful, so it is an ongoing process in areas where less reference sequence is available.

As of today, it remains that these sources of error makes this approach less ideal for single nucleotide variation analysis (e.g. comparing single nucleotide polymorphisms (SNPs)) which is likely to be important for the VM (but that error rate is likely to change over time).

As summarized in the book chapter, Piñar's lab has utilized this "whole microorganism community" approach to try to solve mysteries about artifacts -- such as the identification of a micro-organism that is from Taiwan (plus a bunch of soil organisms) growing on a statue head that had been stolen and its whereabouts while it was missing is unknown.  Does this mean the statue was buried in Taiwan during its missing time?  Maybe.

But this kind of stuff is pretty speculative, in my view.  Trying to apply this to the VM, would a whole micro-organism community analysis reveal anything other than a "Yale Library Rare Book Room" signature community?  Or even, "the shelf that holds Beinecke MS 408" signature community?  Or probably more likely, "the shelf in the Jesuit library that held the Voynich Manuscript before it was bought by Voynich" signature community?    
Who knows?  

But depending on how the investigation was structured and what other manuscripts were tested as comparators/controls, maybe more of the story could be figured out with this information.

For the last installment on the board's series in "Biocodicology -- A Deeper Dive," I'd like to review the following recent paper in depth:

A biocodicological analysis of the medieval library and archive from Orval Abbey, Belgium
Nicolas Ruffini-Ronzani , Jean-François Nieus , Silvia Soncin , Simon Hickinbotham , Marc Dieu , Julie Bouhy , Catherine Charles , Chiara Ruzzier , Thomas Falmagne , Xavier Hermand , Matthew J. Collins  and Olivier Deparis
Published:02 June 2021; Royal Society Open Science, 8(6).

The full publication is available You are not allowed to view links. Register or Login to view..

1.  What are the major research questions Ruffini-Ronzani et al. asking?
a) Can a process be set up that allows for a full biocodicological review of an entire collection of manuscripts efficiently; specifically, the library of the Orval Abbey located in Belgium?

Yes -- they did this, analyzing both parchment manuscripts and charters (legal documents) for species for almost 1500 samples.  Note results were checked using a ZooMS software that I alluded to in an earlier review by You are not allowed to view links. Register or Login to view. -- which has now been named Bacollite.

b) Can the results say anything about the specific species use for parchment production in this geographic area over the time period of the manuscripts and charter documents involved?

Yes, see results below.

2.  What techniques were utilized?  Only ZooMS protein analysis (in other words, parchment species identification, see comments about this scope below).

3.  What were Ruffini-Ronzani et al.'s results?
I'll quote the abstract here:

"Within the genuine production of the Orval scriptorium (26 units), a balanced use of calfskin (47.1%) and sheepskin (48.5%) was observed, whereas calfskin was less frequent (24.3%) in externally produced units acquired by the monastery (92 units). Calfskin was preferably used for higher quality manuscripts while sheepskin tends to be the standard choice for ‘ordinary’ manuscript book production."

and, as compared to charters, 

"Although the five earliest preserved charters are made of calfskin, from the 1230s onwards, all charters from Orval are written on sheepskin."

A visual representation of the overall results in relation to time is available in Figure 2.

[attachment=6184]

So nothing earth-shattering, particularly from the view of the VM as the differing time frame and pretty focused geographical location may mean there is very little of these results that would apply to the VM.

If you're interested, please see the Figures You are not allowed to view links. Register or Login to view. that divide the data up as to geographic origin, number of folios, textual topic, and a "quality" metric formulated by this publication.

TLDR:  This is the first (of hopefully many) biocodicological reviews of a full collection of historic manuscripts and provides a beginning approach to the kind of systemic review that will be needed to have a database of sequences to help pinpoint the likely geographic location of the VM creation (or at least, where the animals that contributed their skin to the parchment were from).

One small disappointment with this work is that DNA analysis (e.g. recordation of the sequences, either autosomal and/or mitochondrial) of the calves, sheep, and goats involved -- that could have been attempted on these samples -- does not seem to have been done.  Whether this was a monetary decision (pretty high likelihood) or a scientific one (e.g. they tried but got all negative results) it is not known, but truly that would would have been the most helpful work to the VM investigation.

It is possible that they saved a portion of the samples and more close DNA analysis will be performed when money is available -- we can hope.  

Finally, I'll note that I believe these samples were collected from 2017-2019 and analysis and publication took until June of 2021, so the work is not particularly quick.  More labs and more money are the only way to speed this up.  However, establishment of process and approach is absolutely needed, which I see as the biggest contribution of this publication to the overall biocodicology field. 

This concludes an overview of what I see as central movements/publications in this very new area as of January 2022 -- and the one certitude is that future interesting work will be coming out.

Please know that I am available for questions on this topic (and yes, bi3mw, I'm going to be adding to your chicken protein thread if I have further thoughts now that this overview is wrapped up).  If you have specific questions or a publication you'd like a closer look at that are going to be going down a narrow focus, I would suggest starting a new thread in the Physical Materials section to make this string an easier point of reference for a general overview as of early 2022.

I will, however, be adding to this thread when I become aware of new publications that have nice possible application to the VM -- or if such a publication is brought to my attention.  I do appreciate understanding that my "Voynich time" is limited and I've got other irons in the fire, so to speak.  No, there is no "theory" directly ready -- but I'm chasing the stuff that interests me most, best described as eliminating possibilities, and will definitely share (or share through my collaborators) if anything interesting enough to publish comes out.

But biocodicology of the VM will always be a high interest and I'm happy to discuss.

I just attended the Vineland Map presentation which showed a highly in depth analysis of the map and its associated manuscripts.

During the talk they mentioned that the group of Collins, Fiddyment, and Teasdale did an analysis of the source animals of the parchment materials for both the map and the related manuscripts and found a variety of both goat and sheep parchment.  Note that it is thought that the substrate materials are considered to be genuine (from the early to mid-1400s) whereas it is the ink of only the map (and some additional writing on both the reverse and at least a few words on the first page of an associated manuscript) are considered suspect.  Specifically, the ink used contains titanium that is in a crystalline form that is only found with modern manufacture.  The additional analysis work that was done confirms the modern source of this ink component, thus dating the map (and some of the reverse writing and a few words on the manuscript) as sometime between the 1920s and the 1950s.

SPOILER:  They didn't have much to say about who would be the likely forger, but it was an interesting talk nevertheless.

I believe this talk will be available for rewatching on the Yale Library website in the near future.

In any case, they confirmed that all DNA and protein analysis done on the map will be published in a future paper.  I will be keeping an eye out for that and will add to this string with a summary of their findings.  I will specifically be focusing on what increases in accuracy may have been developed since the last set of published data for medieval parchment came out and how that could perhaps be applied to the VM.