search

NHMDenmark / DanSpecify

Important files regarding the Danish instance of the Specify database system for collections digitisation and management, plus placeholder for issue tracking. Guidelines, manuals and other kinds of documentations will be gathered on the wiki.
3 stars 2 forks source link

Fix Herpetology Taxon tree #247

Open FedorSteeman opened 12 months ago

FedorSteeman commented 12 months ago

After investigating the taxon tree with @markscherz we realized that something had gone wrong with previous attempts to import the full taxonomy for Reptiles and Amphibia. It seems the best way forward is deleting all those taxa without any objects associated and then reimport the full taxonomy.

FedorSteeman commented 12 months ago

I now deleted all taxa I could from the test database.

If @markscherz could log on to https://specify-test.science.ku.dk/specify/ and approve, I could repeat the exercise in the live db.

markscherz commented 11 months ago

Hey Fedor, still many duplicates in here unfortunately, and the higher taxonomy is really messed up. For instance, there are two Agamidae entries (one under Iguania, one not), each with a Draco. I think there are no duplicated species, only higher levels. There are some bad remnants of the higher taxonomic mess as well. For instance, Dibamia currently only contains Viperidae, but Viperidae is also elsewhere in the tree, and is not at all affiliated with Dibamia. I guess we would really need to rebuild the tree completely from the tips upwards, in order to get the higher taxonomy remotely correct. An option is to do this manually using the Merge and Move tools, but this is incredibly slow. The higher taxonomic levels that are here also do not map to current taxonomy. For instance, there is no Dasypeltinae anymore, either on CoL or on Reptile Database. It is a synonym of Colubrinae. But that can be dealt with later I guess

FedorSteeman commented 11 months ago

Yes I've noticed too, but I felt the remaining problems can be easily (despite the current slowness) be dealt with manually. Super and subtaxa can be gotten rid off and then a fresh and correct taxonomy can be imported and enmeshed with the current state.

I can give it a try if you trust me and think the task is too daunting right now...

markscherz commented 11 months ago

Okay, agreed. Green light from my side.

FedorSteeman commented 11 months ago

On closer thought, maybe it is actually possible to uproot all genera and moved these to the root of the tree, and then see what happens if I import a complete taxonomy including those genera. There's a possibility the existing genera are then incorporated into the new taxonomy, but I'm not sure. I will run some experiments on test and see what happens.

FedorSteeman commented 3 months ago

Agreed with @markscherz that I just removed all unused taxa from the tree.

FedorSteeman commented 3 months ago

Agreed with @markscherz that I just removed all unused taxa from the tree.

Mark communicated that they first need to import the FileMaker data, before doing so.