[tdwg-content] Evidence/Token Class [was: New Darwin Core terms proposed relating to material samples]

[Richard Pyle]

Thanks, Dan.  I changed the subject line because I think your comment:

> tissue converted to DNA extract is still tissue (the tissue is still 
> "material sample" after "destructive sampling", and still part of 
> the event, just as is a photograph of the beast in the museum.

Relates to part of what Steve wrote:

> We defined a class for evidence, but we also considered not having
evidence 
> being an explicit class.  Not defining an explicit Token class would have
simplified 
> the diagram at the bottom of the page - one could just say that there
should be 
> evidence and it should be linked to the resource it documents.  
> Token and THeE/IndividualOrganism are not disjoint classes - the physical
entity 
> can be the evidence if somebody "owns" it and makes it available for
people to examine.  
> However, in DSW, Token and THeE are not synonymous because we allow
evidence 
> to include things that are not physically derived from the entity (e.g.
images, sounds, 
> string data records) in addition to physical specimens.

This is an area that at first seems subtle, but once we actually started
working with data, starts to became more and more intuitive (at least to me,
anyway).

I first started thinking about this seriously during a 2007 meeting of ICZN
Commissioners, when we were contemplating the question of what could/should,
and what could/should not, fulfill the role of a type specimen under the
ICZN Code.

Needless to say, we did not come up with a complete answer to the question,
but the discussion did help frame in my mind the question in much better
detail.

Under the ICZN Code (which differs in a few respects from the ICNafp Code in
this area), the definition of a type specimen s fairly straight-forward:

1) It must be biological material.  An image (e.g., illustration or
photograph) cannot be a type specimen (I believe the ICNafp accommodates
"iconotypes"). Rather, the organism depicted in an image is regarded as
being the type specimen.

2) A type specimen consists of the entire organism. For example, in the case
of a mammal that is captured and separated into its fur, skeleton,
alcoholic-preserved tissues, and tissue samples for DNA analysis, all of
these things together constitute the type specimen.  In other words, the
whole specimen is the type.  This is in contrast to ICNafp, where a single
specimen from a tree (for example) is the type, whereas other specimens
taken from the same tree are isotypes, and the remainder of the tree is not
a type at all.

3) It is understood that the community of organisms inhabiting a type
specimen (e.g., parasites and commensals) do not form part of the type
specimen.

There are other issues as well (especially for ichnotaxa, colonies, and
other specific cases), but the general point is that under the ICZN Code, a
type specimen consists of the sum of all cells that contain the same DNA
material that were part of the same biological individual (i.e., identical
twins, clones, and parthenogenic and other asexually reproducing organisms
potentially contain the same DNA material, but are different "individuals").

Most of the above is pretty straight-forward.  Where it gets a bit squishy
is when you go down below the "tissue sample" level into smaller and smaller
subunits.

Can a single cell containing intact DNA be a type specimen for a
multicellular organism?

Can a single DNA molecule be a type specimen?

Can a single chromosome be a type specimen?

Can a PCR product be a type specimen?

...you can add as many hair-splitting examples as you wish.

Conversations like this let me to draw a distinction between matter that is
actually from an organism, vs. stuff that is representative of the organism
but lacks any material connection to the organism.

Coming back to Dan Janzen's comment; I would conclude that most things that
would full under the term "materialSample" include actually matter from a
biological organism.  A PCR product is borderline, because it presumably
does contain actual matter from the organism, but the vast majority of
matter in the PCR product was not from the organism.  Once you lose all the
matter from the organism, you are no longer talking about the "individual",
but instead are now talking about "Evidence of the Individual".  This
applies to images of the individual (whether taken in-situ, or in a lab),
DNA "sequences" (rendered as text), images of the results of DNA sequencing
analysis (in various forms), and various other non-biological derivatives.

So, this leads me to the comment from Steve about "Evidence" (which, if I
understand correctly, is conceptually the same as the DSW "Token":
https://code.google.com/p/darwin-sw/wiki/ClassToken).  In our model, a
specimen is *not* a Token or Evidence -- it is an individual.  This has
proven (in my mind at least -- Rob W. still has reservations) to be a MUCH
more practical way to think of, and manage data for, specimens.  

This email will be WAY too long if I try to explain all the subtleties of
this, so I'll just leave it at that.  But clearly there are some differences
between our notion of "Evidence", and the DSW concept of "Token" -- although
both seem to be intended to solve the same problem.

Enough for now....

Aloha,
Rich