search

globalbioticinteractions / refuted-biotic-interactions-by-eol

Biotic interaction data that failed Encyclopedia of Life data validation
0 stars 0 forks source link

Sponge Hononym in Pollination interaction #3

Open jtmiller28 opened 2 years ago

jtmiller28 commented 2 years ago

There is a homonym for Chondrilla juncea that globi is interpreting as a member of the sponges within the pollinator interaction data. The current path is source: Animalia | Arthropoda | Insecta | Hymenoptera | Apoidea | Halictidae | Lasioglossum target: Animalia | Porifera | Demospongiae | Chondrosida | Chondrillidae | Chondrilla

The correct path for the target should be: Plantae | Tracheophyta | Magnoliospida | Asterales | Asteraceae | Chondrilla L. | Chondrilla juncea L.

ITIS Taxonomy: https://www.itis.gov/servlet/SingleRpt/SingleRpt?search_topic=TSN&search_value=37029#null

Thanks!

seltmann commented 2 years ago

@jtmiller28 and I are wondering how best to provide information to other GloBI users, such as a refute for the interaction or is there a way to prioritize one taxon over another?

jhpoelen commented 2 years ago

@jtmiller28 thanks for sharing your observations about the homonym Chondrilla juncea .

Can you please point to the specific claim that included the suspicious interaction?

jhpoelen commented 2 years ago

From https://www.globalbioticinteractions.org/?interactionType=interactsWith&sourceTaxon=Chondrilla%20juncea&targetTaxon=Lasioglossum , I found the following records:

suspicious-sponge-interactions.csv

but . . . for some reason, I am not seeing the

Lasioglossum (genus) <> Chondrilla (sponge genus) interaction.

Can you please help me find your suspicious interaction?

Screenshot from 2022-07-11 12-09-21

jhpoelen commented 2 years ago

@jtmiller28 oops, never mind. I think I might have found the suspicious interaction via

https://www.globalbioticinteractions.org/?interactionType=interactsWith&sourceTaxon=Chondrilla%20&targetTaxon=Lasioglossum

My earlier search was searching for the specific Chondrilla species, not the genus. This excluded the suspicious interaction you alluded to. Screenshot from 2022-07-11 12-12-47

From the search results, there's a visual hint to it being a homonym because two ITIS logos are present under the name. These ITIS icons point to:

the plant genus http://www.itis.gov/servlet/SingleRpt/SingleRpt?search_topic=TSN&search_value=37028 Screenshot from 2022-07-11 12-14-07

and the sponge genus http://www.itis.gov/servlet/SingleRpt/SingleRpt?search_topic=TSN&search_value=48719 Screenshot from 2022-07-11 12-14-54

jhpoelen commented 2 years ago

Tracing the provenance of the claim via a recent web-of-life GloBI review - 

$ curl "https://depot.globalbioticinteractions.org/reviews/globalbioticinteractions/web-of-life/indexed-interactions.tsv" | grep "Chondrilla" | grep "Lasioglossum"
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100 60882    0 60882    0     0   112k      0 --:--:-- --:--:-- --:--:--  112khttps://en.wiktionary.org/wiki/support    Chondrilla juncea                                           http://purl.obolibrary.org/obo/RO_0002456   pollinatedBy                            Lasioglossum malachurum                                     38.014466   23.635043                   Petanidou, T. (1991). Pollination ecology in a phryganic ecosystem. Unp. PhD. Thesis, Aristotelian University, Thessaloniki.    globalbioticinteractions/web-of-life    Web of Life. http://www.web-of-life.es .    https://github.com/globalbioticinteractions/web-of-life/archive/f63e4dbd52d31b698df5d5f63a409889a26fd0a5.zip    2022-07-02T14:03:58.433Z    794e639174dfa96223fea4aeba6ad7f93d873b66275d5712c5db689f25541856    0.12.4
https://en.wiktionary.org/wiki/support                          Chondrilla juncea                           http://purl.obolibrary.org/obo/RO_0002456   pollinatedBy                            Lasioglossum nigripes           38.014466   23.635043                   Petanidou, T. (1991). Pollination ecology in a phryganic ecosystem. Unp. PhD. Thesis, Aristotelian University, Thessaloniki.    globalbioticinteractions/web-of-life    Web of Life. http://www.web-of-life.es .    https://github.com/globalbioticinteractions/web-of-life/archive/f63e4dbd52d31b698df5d5f63a409889a26fd0a5.zip    2022-07-02T14:03:58.433Z    794e639174dfa96223fea4aeba6ad7f93d873b66275d5712c5db689f25541856    0.12.4
https://en.wiktionary.org/wiki/support                          Chondrilla ramosissima                  http://purl.obolibrary.org/obo/RO_0002456   pollinatedBy                            Lasioglossum limbelloides       38.014466   23.635043                   Petanidou, T. (1991). Pollination ecology in a phryganic ecosystem. Unp. PhD. Thesis, Aristotelian University, Thessaloniki.    globalbioticinteractions/web-of-life    Web of Life. http://www.web-of-life.es .    https://github.com/globalbioticinteractions/web-of-life/archive/f63e4dbd52d31b698df5d5f63a409889a26fd0a5.zip    2022-07-02T14:03:58.433Z    794e639174dfa96223fea4aeba6ad7f93d873b66275d5712c5db689f25541856    0.12.4
https://en.wiktionary.org/wiki/support                          Chondrilla ramosissima                  http://purl.obolibrary.org/obo/RO_0002456   pollinatedBy                            Lasioglossum limbellum          38.014466   23.635043                   Petanidou, T. (1991). Pollination ecology in a phryganic ecosystem. Unp. PhD. Thesis, Aristotelian University, Thessaloniki.    globalbioticinteractions/web-of-life    Web of Life. http://www.web-of-life.es .    https://github.com/globalbioticinteractions/web-of-life/archive/f63e4dbd52d31b698df5d5f63a409889a26fd0a5.zip    2022-07-02T14:03:58.433Z    794e639174dfa96223fea4aeba6ad7f93d873b66275d5712c5db689f25541856    0.12.4
https://en.wiktionary.org/wiki/support                          Chondrilla ramosissima                  http://purl.obolibrary.org/obo/RO_0002456   pollinatedBy                            Lasioglossum malachurum         38.014466   23.635043                   Petanidou, T. (1991). Pollination ecology in a phryganic ecosystem. Unp. PhD. Thesis, Aristotelian University, Thessaloniki.    globalbioticinteractions/web-of-life    Web of Life. http://www.web-of-life.es .    https://github.com/globalbioticinteractions/web-of-life/archive/f63e4dbd52d31b698df5d5f63a409889a26fd0a5.zip    2022-07-02T14:03:58.433Z    794e639174dfa96223fea4aeba6ad7f93d873b66275d5712c5db689f25541856    0.12.4
https://en.wiktionary.org/wiki/support                          Chondrilla ramosissima                  http://purl.obolibrary.org/obo/RO_0002456   pollinatedBy                            Lasioglossum villosulum         38.014466   23.635043                   Petanidou, T. (1991). Pollination ecology in a phryganic ecosystem. Unp. PhD. Thesis, Aristotelian University, Thessaloniki.    globalbioticinteractions/web-of-life    Web of Life. http://www.web-of-life.es .    https://github.com/globalbioticinteractions/web-of-life/archive/f63e4dbd52d31b698df5d5f63a409889a26fd0a5.zip    2022-07-02T14:03:58.433Z    794e639174dfa96223fea4aeba6ad7f93d873b66275d5712c5db689f25541856    0.12.4
100 42.7M    0 42.7M    0     0  15.3M      0 --:--:--  0:00:02 --:--:-- 15.3M

Noting to no explicit genus was found for neither geni .

jhpoelen commented 2 years ago

So, it turns out the issue @jtmiller28 found has it's origin in a generalizing species -> genus mapping.

$ curl "https://zenodo.org/record/6394935/files/taxonMap.tsv.gz" | gunzip | grep "Chondrilla ramosissima"
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
 22 33.9M   22 7951k    0     0  2013k      0  0:00:17  0:00:03  0:00:14 2013k  Chondrilla ramosissima  EOL:43003   Chondrilla
    Chondrilla ramosissima  GBIF:2248823    Chondrilla
    Chondrilla ramosissima  GBIF:6304786    Lactuca viminea viminea
    Chondrilla ramosissima  GBIF:7945704    Chondrilla ramosissima
    Chondrilla ramosissima  IRMNG:1283520   Chondrilla
    Chondrilla ramosissima  ITIS:48719  Chondrilla
    Chondrilla ramosissima  WORMS:1095160   Chondrilla ramosissima
    Chondrilla ramosissima  WORMS:1138913   Chondrilla ramosissima
    Chondrilla ramosissima  WORMS:132021    Chondrilla
100 33.9M  100 33.9M    0     0  3934k      0  0:00:08  0:00:08 --:--:-- 5490k

Which then propagated the idea that Chondrilla ramosissima ITIS:48719 Chondrilla which then hit the homonym issue.

jtmiller28 commented 2 years ago

Yep that's the one, for future reference when "pointing" to to the interaction, should I search for it on Globi and post the link? Or is there a ID associated from the data itself that I should include?

jhpoelen commented 2 years ago

@jtmiller28 thanks for confirming.

What I usually do is provide pointers to suspicious interactions is:

  1. search url
  2. a screenshot search results and
  3. a sample csv downloaded from the "sample csv" search page on globalbioticinteractions.
jtmiller28 commented 2 years ago

Gotcha, I'll use this format for future issues. Thank you!

jhpoelen commented 2 years ago

PS I've created an list of likely species -> genus mappings defined in GloBI's taxon graph at:

https://github.com/jhpoelen/globi-taxon-graph-species-to-genus-mapping

and used a name-alignment template to pick up the NONE matches.

@jtmiller28 you've probably helped uncover a whole slew of species names that were "downgraded" to genus names, likely due to optimistic mapping rules that are no longer in use (e.g., if species name doesn't match, try the genus instead), but have left their traces in datasets.

Question for you and @seltmann - when a species name is unknown or invalid (e.g., Homo sapienz) what you like Nomer and GloBI to do? Map to genus? Or just leave them as is, and not link them to any known taxonomic hierarchy (e.g. ,itis, gbif, ncbi)?

Curious to hear your thoughts. . .

jtmiller28 commented 2 years ago

Hmm we seem to lose resolution in either case, from my research I would think that mapping to genus as the stop point would be the safer option since generating large lists of thousands of species makes it difficult to locate misspellings. Is there a step that could generate a warning if genus is used as a stop point due to unfound species in the catalogue that might be ideal? If a warning is appended to the end of the query then the user can manually locate misspellings to resolve.

@seltmann curious as to what your opinion is on this as well.