search

OpenTreeOfLife / hackathon

A repo for the 2014 OpenTree / Arbor / HIP hackathon
8 stars 1 forks source link

Integrate OpenTree with Interaction Webs #13

Open bobthacker opened 10 years ago

bobthacker commented 10 years ago

Is it possible to map host-symbiont associations onto OpenTree? Can resources such as the interaction web database be used to query OpenTree? http://www.nceas.ucsb.edu/interactionweb/resources.html

sckott commented 10 years ago

For interaction webs, you may be interested in hooking in with @tpoisot mangal, and there's an R library too

tpoisot commented 10 years ago

I would be super happy to help. If we are lucky funding wise, we'll be able to import the whole IWDB from NCEAS into mangal this summer. Alternatively, if you have unique identifiers for the taxa, I can add that in the data specification, and then searching the database would be extremely simple. I also have a python module, which is better than the R version to do rapid development (in my opinion).

jhpoelen commented 10 years ago

Interesting! You might be interested in looking at http://globalbioticinteractions.org . We have an open issue for integrating IWDB (https://github.com/jhpoelen/eol-globi-data/issues/65). Also, we have almost 20 others sources of species interactions data. @sckott helped us with creating a R library for it too. Please let me know if you have any questions.

Seems like mangal and GloBI are doing pretty similar things! Perhaps we should join forces. . .

jhpoelen commented 10 years ago

This issue seems closely related to #9 .

tpoisot commented 10 years ago

@jhpoelen sure - but the focus of mangal is slightly different as we put much emphasis on species traits, and other metadata, as I think these are needed informations to really go beyond a catalogue of species interactions and into predictive things. But writing a wrapper for the two DB to be queried by the same package should be feasible. We have a quite long data specification, see http://biorxiv.org/content/early/2014/02/12/002634 for more infos. Let's move the discussion elsewhere, but it would be interesting (I think) to aggregate both DB at the same place.

mjy commented 10 years ago

Might be of interest- the biological associations model being developed in TaxonWorks seems to match the specification in mangal pretty nicely. Among other things it allows for domain/range properties to be asserted on relationships, and relationships to be asserted between taxa or collection objects (= populations or individuals). Code wise we have the basic models (e.g. https://github.com/SpeciesFileGroup/taxonworks/blob/master/app/models/biological_association.rb), but no interface for that aspect, but we're rapidly scaffolding things right now.

jhpoelen commented 10 years ago

@tpoisot sounds good. I'll contact you offline about this.

For completeness, I do want to say that GloBI contains meta-data or traits related to the interaction such as: life stage, body parts, habitat/environment, location, depth / altitude, and ecoregions. The majority of terms are coded using existing ontologies such as RO, EnvO, Uberon and taxonomy providers like WoRMS, ITIS, NCBI, EOL, GBIF. In addition, the references related to the interaction observations are available. The use of coded terms allows for integration of rich datasets (e.g. more species traits, occurrence data, phylogenes, genomics) from sources like EOL, GBIF, and OpenTree (!) to allow for fancy systems prediction modeling or cross-disciplinary analysis.

jhpoelen commented 9 years ago

In preparation for the hack-a-thon upcoming week, I took a stab at integrating OpenTree Taxonomy into GloBI. Now OT Taxon ids can be used to query biotic interactions. For instance: http://api.globalbioticinteractions.org/taxon/OTT%3A770315/hostOf provides a list of organisms that a human (Homo sapiens) hosts. I am hoping that this will open the door to a more appealing integration to allow for phylogenetic comparison of interacting organisms.

bobthacker commented 9 years ago

In the GloBI api -- I saw the set of interaction types like "hostOf" and "parasiteOf"

Can these be modified to include more general symbiosis, e.g. "symbiontOf" ?

In many cases, "parasite" (negative) or "mutualist" (positive) status is not actually known.

On Fri, Sep 12, 2014 at 11:14 AM, jhpoelen notifications@github.com wrote:

In preparation for the hack-a-thon upcoming week, I took a stab at integrating OpenTree Taxonomy into GloBI. Now OT Taxon ids can be used to query biotic interactions. For instance: http://api.globalbioticinteractions.org/taxon/OTT%3A770315/hostOf provides a list of organisms that a human (Homo sapiens) hosts. I am hoping that this will open the door to a more appealing integration to allow for phylogenetic comparison of interacting organisms.

— Reply to this email directly or view it on GitHub https://github.com/OpenTreeOfLife/hackathon/issues/13#issuecomment-55425303 .

Robert W. Thacker, PhD Professor Department of Biology University of Alabama at Birmingham 464 Campbell Hall 1300 University Boulevard Birmingham, AL 35294-1170 voice: 205-934-9685 fax: 205-975-6097 email: thacker@uab.edu http://www.uab.edu/cas/biology/thacker http://www.uab.edu/biology/thacker http://www.portol.org

jhpoelen commented 9 years ago

Please see https://github.com/lukejharmon/traitathon/wiki/Pocket-Gophers-and-Their-Parasitic-Chewing-Lice for a specific example on how to integrate GloBI interaction data with Open Tree of Life phylogenetic trees. Please re-open issue if you'd like to share thoughts or have questions.