CatalogueOfLife / xcol

Working towards the extended Catalogue of Life Checklist
0 stars 0 forks source link

Investigate rank order invalid taxa #131

Open mdoering opened 3 weeks ago

mdoering commented 3 weeks ago

https://www.checklistbank.org/dataset/298754/names?facet=rank&facet=issue&facet=status&facet=nomStatus&facet=nomCode&facet=nameType&facet=field&facet=authorship&facet=authorshipYear&facet=extinct&facet=environment&facet=origin&facet=sectorMode&facet=secondarySourceGroup&facet=sectorDatasetKey&facet=group&issue=classification%20rank%20order%20invalid&limit=50&offset=0&sectorMode=merge

DianRHR commented 2 weeks ago

There are different cases:

  1. family Ancistrocomidae in its original source is placed under Kingdom Chromista | Phylm Ciliophora | class Phyllopharyngea | Subclass Rhynchodia | Order Rhynchodida. The Order Rhynchodida is not included in COL but Rhynchodia is a synonym plant genus that was merged in the xrelease. The code matched Rhynchodia and placed a ciliophoran family under the accepted plant genus of Rhynchodia.
    Class Phyllopharyngea is in COL, therefore Ancistrocomidae should be merged as an unplaced family under Phyllopharyngea. The sector in the has well defined the subject and the target.

  2. The fish genus Hemiloricaria in its original sources WoRMS and ITIS is under Loricariidae | Loricariinae. Loricariinae is a synonym unranked name in COL related to the accepted plant genus Antennariinae (family Asteraceae) and the genus was erroneously merged there. The genus should be merged under the fish family Loricariidae. Both sectors have well defined the subject and the target. The genus was merged twice eventhough the taxonomy is the same in both sectors. However, Antennariinae and Loricariinae (Plants) are not included in the COL 24.

  3. 48 Species of 'Persica' merged under Prunus amigdalus. Persica in its original source is synonym of Prunus amigdalus. Therefore, it causes that all species of Persica are being merged below Prunus amigdalus. The genus 'Persica' is not in COL, and two sources (WFO and WCVP) consider it as synonym of the species Prunus amigdalus. In WFO those species of Persica are "unchecked names" as the example of Persica accensa.

DianRHR commented 1 week ago

More examples:

  1. Family Ceratoporellidae was merged under family Helioporidae. In the original source this family is below order Helioporaceae, but that's an unaccepted order name and its accepted name is Helioporidae (according to WoRMS) . Family Periscyphidae has the same situation. Based on that logic the merge is working fine, but still merging a family under another family.

  2. Families Silphidae and Scydmaenidae were merged under Family Staphylinidae eventhogh both of them are under superfamily Staphylinoidea, which is already in the baseCOL. However, in the base COL the family Staphylinidae includes the superfamily Staphylinoidea Ganglbauer, 1895 as synonym. That might be the reason why those families are merging wrongly under another family , instead of merging under the accepted superfamily Staphylinoidea Latreille, 1802.

mdoering commented 5 days ago

An interesting problem. What would we want to see as an outcome?

We currently get this, as Ceratoporellidae is underneath order Helioporacea in ITIS:

Scleralcyonacea McFadden, van Ofwegen & Quattrini, 2022 [order]
  Helioporidae Moseley, 1876 [family]
    =Helioporacea Bock, 1938 [order]
    Ceratoporellidae [family]
    Heliopora de Blainville, 1830 [genus]

Should we rather:

    • make family Ceratoporellidae a synonym of Helioporidae?
    • simply ignore the family and place all children under Helioporidae?
    • stop processing the family and all its descendants in such a source?
    • keep the family, but put it next to Helioporidae and below the order Scleralcyonacea
mdoering commented 5 days ago

I think I would go option 1 or 4, but slightly prefer the later. I have implemented that for now!

mdoering commented 4 days ago

I have deployed an improvement that should follow option 4. Please verify the outcome of the new weekend build and close this issue if it's addressed!

DianRHR commented 1 day ago

That solution is pragmatical and it worked for cases: 2, 3, 4, 5

However: Case 1. Ancistromidae is still merging erroneously under the plant Order Gentianales. Based on the higher taxonomy in the origial source, it should be merging under class Phyllopharyngea (Phylum Ciliophora)