vector role (cloning and disease)

[obi-bot]

see dev call 27/08/2008 notes

discussion copied below. AI is to define properly the terms cloning vector role and disease vector role. maybe also having a parent vector role.

Daniel Schober: a role of a circular DNA or RNA sequence to serve as a carrier of genetic information during cloning Daniel Schober: is it not self connected? Chris Stoeckert: not sure what you mean by self connected Daniel Schober: I mean the circular structure can fold in differrent forms? Chris Stoeckert: lambda is linear as I recall Daniel Schober: probably add that a cloning vector has known restriction sites (i.e. must be fully sequenced...? Daniel Schober: it serves as matric / carrier of genetic information to be investigated further. Daniel Schober: there are yeast artificial chromosomes. Do they serve as vectors? Chris Stoeckert: YACs are vectors! Chris Stoeckert: but YACS are not real chromosomes Daniel Schober: on google, do define: cloning vector, get many results. Chris Stoeckert: viral vectors Melanie Courtot: viral you mean like disease vectors right? I will create vector and specify disease and cloning - def to be worked on both - seems ok? Chris Stoeckert: have the oxford dictionary ofbiochemistry and molecular biology Daniel Schober: http://www.accessexcellence.org/RC/VL/GG/YAC.php so a vector can be of chromosomal origin. Chris Stoeckert: cloning vector is the DNA of any transmissable agent into which a foreign DNA can be spliced (I would say inserted) Daniel Schober: yes, the important intrinsic property is that it has known restriction sites and is propagated when the organism propagates. Chris Stoeckert: of chromosomal origin but these are not chromosomes. they are the ARS signal from chromomsomes basically and nothing else Daniel Schober: yes, DNA or RNA fragment.

Reported by: mcourtot

Original Ticket: "obi/obi-terms/102":https://sourceforge.net/p/obi/obi-terms/102

[obi-bot]

Logged In: YES user_id=138355 Originator: NO

I have my usual concern about "vector role" perhaps being too general. I'd suggest building the more specific ones, starting with cloning vector role with the definition that Chris gives, with Joe's email comments considered.

Minor nit: Daniel "important intrinsic property is that it has known restriction sites." That there are restriction sites is intrinsic. That they are known is not.

Original comment by: alanruttenberg

[obi-bot]

The way the term vector is used in the life-sciences, is that one material entity transfers _some_ material entity to another material entity. I am sure you are aware of the examples, but the use can vary from whole organisms (a vector for disease) to plasmids and viruses acting as vectors for specific sequences of DNA. In most cases we want to be able to say that a particular material entity is a vector, and that it is a vector for another particular type of entity If we take the example used above of the viral vector then it is possible we want to make the following statement:

Virus has_role some vector_role and is_vector_for some DNA construct

or more generically

material_entity has_role some vector and is_vector_of some material_entity

Every single term used in the ontology should be defined in the ontology. Regarding the inclusion of biological_vector_role, this means that some where in the ontology we need a class for biological, vector and role. During the bio-material discussion trying to describe what biological means, proved to be very difficult, so difficult that it is not in the ontology. I would suggest not using the word biological (bio-) for any of the class names until bio- or biological are classes in there own rights.

I understand that just using the term vector can be confusing as it has alternative meanings in maths and and computer science. Therefore I would suggest using the term material_vector_role instead.

I think the addition of specific types of material_vector_roles are not ideal and will cause alot of terms to be added to the ontology uneccessarily and you have to conflate the role vector and what it is a vector for into one class asserted under role.

For example, the issue surrounding cloning vector is not actually to create the class cloning_vector_role, rather the question was, how do we describe a cloning vector?

To describe the class cloning_vector I would suggest the following: That a cloning_vector is only DNA which bears the material_vector_role and is a material_vector_for some DNA

This avoids the need to create specific vector_roles for every is_vector_for something.

In the cloning example you could expand on the is a material_vector_for some DNA and describe the _some_DNA_. For example, you could describe the some DNA as the DNA is_gene_encoding_ for E. coli β-galactosidase.

I hope this presents a more efficient and scalable mechanism to deal with a material vectors and what it is a vector for. If anything, please stop using the strings bio- or biological in class labels :)

Cheers

Frank

Original comment by: fgibson_sf

[obi-bot]

Ryan Brinkman: I have used vector in the computer science / math sense for a term in OBI in the data transformation branch and it would be good to be able to distinguish between these two, in as suggested by CS

Original comment by: rbrinkman

[obi-bot]

Paraphrasing Dr. Math (how can you not believe someone named Dr. Math!): A vector is a mathematical concept that is described by quantities with magnitude and direction and for which the properties of commutivity and associativity hold. http://mathforum.org/library/drmath/view/55463.html

Chris

Original comment by: cstoeckert

[obi-bot]

To respond to Frank's proposal, I agree in spirit but would simply label the term as vector role (not material vector role). Vector in the mathematical sense is strictly defined and not a role (see definition proposed in last comment). Thanks, Chris

Original comment by: cstoeckert

[obi-bot]

The issue is that under BFO the only think that can bear a role are things in reality. According to BFO, mathematical concepts are not reality and therefore can not bare roles. - don't shoot the messenger, this is not my opinion :)

If we could deal directly with material entities bearing a vector role in the fist instance we may be able to make some progress on it. The class label I suggested was material_entity_vector. In theory this would allow a mathematical_entity_vector_role BFO permitting......

Original comment by: fgibson_sf

[obi-bot]

OK, I hit submit before seeing Chris's last post - no issue on mathematical vector role then

Original comment by: fgibson_sf

[obi-bot]

• status: open --> closed

Original comment by: jenfostel

[obi-bot]

please dont close a tracker item without posting the solution - As far as I am aware the issues have not been addresses

Original comment by: fgibson_sf

[obi-bot]

• status: closed --> open

Original comment by: fgibson_sf

[obi-bot]

just checked the definition of role and a role can inhere in a continuant entity, not necessarily only independent continuant entity. therefore we could have both mathematical and biological vectors in OBI. I use biological to mean pertaining to biology or to life and living things (wordnet.princeton.edu/perl/webwn). OBI members suggested this term; we can replace if it is confusing or imprecise.

current definition of biological vector: a biological vector role is a role which inheres in a material entity and is realized by the process of transmitting material to the organism that is the target of the transmission.

A cloning vector role is a vector role played by a small, self-replicating DNA or RNA molecule - usually a plasmid or chromosome - and realized in a process whereby foreign DNA or RNA is inserted into the vector during the process of cloning.

we are waiting for processes to be defined in order to complete this definition.

Original comment by: jenniferfostel

[obi-bot]

check out tracker item 1897873. can we close this one as well?

Original comment by: jenniferfostel

[obi-bot]

Nearly there. Can you try and make a point of following the thread, from the bottom to the top. The proposed solution is there. Can we drop the word biological please. We keep having this discussion, from trying to define "biomaterial" we have not worked out how to define "living" or "biological" or "natural" within OBI therefore, these strings shoudl appear nowhere withing the ontology.

I am re-adding this comment from earlier. Jennifer can you please comment on this proposal by either accepting it and making the changes or comment point-by-point where you disagree and offer an alternative, please.

The way the term vector is used in the life-sciences, is that one material entity transfers _some_ material entity to another material entity. I am sure you are aware of the examples, but the use can vary from whole organisms (a vector for disease) to plasmids and viruses acting as vectors for specific sequences of DNA. In most cases we want to be able to say that a particular material entity is a vector, and that it is a vector for another particular type of entity If we take the example used above of the viral vector then it is possible we want to make the following statement:

Virus has_role some vector_role and is_vector_for some DNA construct

or more generically

material_entity has_role some vector and is_vector_of some material_entity

Every single term used in the ontology should be defined in the ontology. Regarding the inclusion of biological_vector_role, this means that some where in the ontology we need a class for biological, vector and role. During the bio-material discussion trying to describe what biological means, proved to be very difficult, so difficult that it is not in the ontology. I would suggest not using the word biological (bio-) for any of the class names until bio- or biological are classes in there own rights.

I understand that just using the term vector can be confusing as it has alternative meanings in maths and and computer science. Therefore I would suggest using the term material_vector_role instead.

I think the addition of specific types of material_vector_roles are not ideal and will cause alot of terms to be added to the ontology uneccessarily and you have to conflate the role vector and what it is a vector for into one class asserted under role.

For example, the issue surrounding cloning vector is not actually to create the class cloning_vector_role, rather the question was, how do we describe a cloning vector?

To describe the class cloning_vector I would suggest the following: That a cloning_vector is only DNA which bears the material_vector_role and is a material_vector_for some DNA

This avoids the need to create specific vector_roles for every is_vector_for something.

In the cloning example you could expand on the is a material_vector_for some DNA and describe the _some_DNA_. For example, you could describe the some DNA as the DNA is_gene_encoding_ for E. coli β-galactosidase.

I hope this presents a more efficient and scalable mechanism to deal with a material vectors and what it is a vector for. If anything, please stop using the strings bio- or biological in class labels :)

Original comment by: fgibson_sf

[obi-bot]

From Wikipedia: There are two types of vectors that convey infectious organisms to a host: mechanical and biological. Microbes do not multiply within mechanical vectors: Mechanical vectors just physically transport microbes from host to host, whereas, within the biological vector, the microbes propagate before the biological vector can transmit the microbes. The OBI term is in conflict. We will discuss changing to material vector as a substitute in the next call.

Original comment by: jenfostel

[obi-bot]

Jennifer, please read the thread and comment on the proposal. The text from wikipedia is irrelevant. There is absolutely no need to introduce terms like mechanical vector and biological vector. Never use the word biological. Please read the thread and the proposed solution!

Original comment by: fgibson_sf

[obi-bot]

• status: open --> closed

Original comment by: bpeters42

[obi-bot]

Check if ready to close with Frank

Original comment by: bpeters42

[obi-bot]

OK, I just read through from start to finish. Agree that it should not be closed. My take: - I like the label material vector role - instead of a 'material vector for' relation, I would go with an approach similar to the 'cloning vector insert role', that Kevin and I had been using. I don't have file access right now, need to check if I am using the right labels here. basically, there is a cloning process with an input entities bearing the cloning vector role and cloning insert role and an output entity recombinant vector. - I think I would like to keep the children cloning vector role and others, but could be convinced otherwise - I think 'role' is problematic, as several 'cloning vector roles', should be 'cloning vector functions'. This may be an ideal scenario to try out the 'capacity' approach, namely using 'material vector capacity' as a child of reaiizable entity, with 'role of' or 'function of' to distinguish the natural and manufactured cases. We don't have to do this now (before the 1.0 release) though, as we would probably have to revisit a bunch of other terms in parallel.

Original comment by: bpeters42

[obi-bot]

I think it should remain open until resolved on this thread.

Original comment by: bpeters42

[obi-bot]

• priority: 5 --> 8
• status: closed --> open

Original comment by: bpeters42

[obi-bot]

Ok, so lets deal with the role issue. A recombinant DNA molecule "can be" a vector, which I think what your were tryign to get across with capacity. I think this is the definition of disposition. However, I think we would agree that it is not a role, but a capacity/disposition - lets agree that it is a realizable_entity for now.

I think we are actually agreeing on the same thing but I am asking for more information, which I dont beleive can be answered with your approach. The relation would be in addition to your structure not in replace of it. The reason for the relation is to answer the question, what is this a vector for , or what is the cloning vector for? If you think of the following examples and how you could answer the question without the relation.

A cloning vector is a vector for DNA construct, A mosquito is a vector_for protozoan parasites

To use your recombinant_plasmid_cloning process example, the relation would allow your to say that the input material plasmid is_vector_for DNA_fragment which is actually the key piece of information missing from the restrictions in the file.

For simplicity we could ignore the concept of vector_realisable_entity class and create a defined class that was a material that is_vector_for some material. Then all these physical materials that are vectors could be infered underneath it

Original comment by: fgibson_sf

[obi-bot]

there are now clear vector roles for 'cloning vector role' and 'biological vector role' - has this been resolved therefore?

thanks Helen

Original comment by: helenp

[obi-bot]

The point is that there should not be cloning vector role and biological role anywhere in the ontology, and we require an additional relation that actualy states what is is a vector for - what material entity is the vector transfering - so no not resolved yet. :) I can just go ahead and fix this

Original comment by: fgibson_sf

[obi-bot]

• assigned_to: jenfostel --> parkinso

Original comment by: helenp

[obi-bot]

• priority: 8 --> 7

Original comment by: bpeters42

[obi-bot]

see tracker items 2902569 - request to take out the word biological from biological vector (making it just vector role) - and tracker 2902583 - request to add logical def for cloning vector. currently no use case for disease vector, but we can consider adding it if someone requests it.

Original comment by: jenfostel

[obi-bot]

Dev call, Jan 25, 2010 JF suggest to lower the priority of role related terms to 5. Would like to discuss all the role terms at the same time.

Original comment by: zhengj2007

[obi-bot]

• priority: 7 --> 5

Original comment by: zhengj2007

[obi-bot]

• assigned_to: Helen Parkinson --> jzheng

Original comment by: bpeters42

[obi-bot]

Discussed on call 6/29/2015.

This issues is still contentious. It is clear that 'cloning vectors' are something that belongs in OBI. The current modeling of the 'cloning vector role' is perfectly fine, and it is a good child of 'material to be added role'. But the modeling of 'biological vector role' is outside of the scope of OBI if it is meant to cover disease vectors in the wild and the current modeling is incorrect (as the only realization of this role would be in protocols, but the definition implies that this could also happen in the wild.

As this has been dragging on for so long, proposal to deprecate OBI_0000109, 'biological vector role' from OBI. 'vector organism' as relevant to infectious diseases should be defined in IDO.

Jie to deprecate, and submit to IDO.

Original comment by: bpeters42

[obi-bot]

• status: open --> closed

Original comment by: zhengj2007

[obi-bot]

'biological vector role' is deprecated. see: SVN revision 3992

The tracker to IDO has been submitted https://code.google.com/p/infectious-disease-ontology/issues/detail?id=42

Original comment by: zhengj2007