🐛❗ BIII dump fix & update

[matuskalas]

https://github.com/research-software-ecosystem/content/blame/0bd520615dd9e4c0d3dd46089f321ab75ec7e68f/datasets/bioschemas-biii-dump.ttl#L4

Hey @albangaignard @hmenager et al.,

Could you please:

• Fix sc:applicationCategory to include the has topic from BIII instead of has function (which is correctly in the featureList)
• Update the dump with the current state of BIII

I'm very sorry, but this is urgent now, as I need all that information for the crucial decisions within EDAM Bioimaging (to see what concepts are used in BIII).

Million giant thanks! 🙏🏽❤

[albangaignard]

Thanks @matuskalas for identifying the issue. I fixed it, it was a silly typo. And many thanks to @hmenager for helping me with the github actions ! I generated a fixed sampe file (10 entries). The whole dataset will be ready within one hour.

[albangaignard]

Here is an fixed entry (RDF turtle):

<https://dev.biii.eurobioimaging.eu/dextrusion> a ns1:SoftwareApplication ;
    ns1:applicationSubCategory <http://edamontology.org/topic_Fluorescent_microscopy> ;
    ns1:citation <https://doi.org/10.1242/dev.201747>,
        "DeXtrusion: automatic recognition of epithelial cell extrusion through machine learning in vivo . Alexis Villars  , Gaëlle Letort , Léo Valon, Romain Levayer Development (2023) 150 (13): dev201747." ;
    ns1:dateCreated "2024-10-24T13:56:22+00:00" ;
    ns1:dateModified "2024-10-24T14:11:22+00:00" ;
    ns1:description """DeXtrusion is a machine learning based python pipeline to detect cell extrusions in epithelial tissues movies. It can also detect cell divisions and SOPs, and can easily be trained to detect other dynamic events.
DeXtrusion takes as input a movie of an epithelium and outputs the spatio-temporal location of cell extrusion events or other event as cell divisions. The movie is discretized into small overlapping rolling windows which are individually classified for event detection by a trained neural network. Results are then put together in event probability map for the whole movie or as spatio-temporal points indicating each event.
""" ;
    ns1:featureList <http://edamontology.org/operation_0226>,
        <http://edamontology.org/operation_Object_detection> ;
    ns1:license "BSD 3" ;
    ns1:name "DeXtrusion" ;
    ns1:publisher "Letort Gaëlle orcid.org/0000-0002-5866-0322" .

[albangaignard]

To stay in line with the Bioschemas ComputationalTool profile, we use now schema:applicationSubCategory to refer to the EDAM topics.

[matuskalas]

Quadrillion thanks @albangaignard & @hmenager !!! 🙏🏽❤

To stay in line with the Bioschemas ComputationalTool profile, we use now schema:applicationSubCategory to refer to the EDAM topics.

Indeed, now I remember that it was I who defined it that way ages ago at Berges de Seine 😅 And appCategory should have been something like "Scientific computational tool"

[matuskalas]

Sorry, still a bug: it only exports one topic if there are multiple.

And an independent question: How difficult would it be to also make a full dump in the native format/BISE ontology, as an alternative to the Bioschemas one? It's not urgent, but would be super nice to have at some point relatively soon, to be able to see ALL the info recorded in BIII, in one file.

[albangaignard]

Hi @matuskalas, I checked and it should be fixed now (https://github.com/research-software-ecosystem/utils/pull/33/files).

For your second question, it's not difficult at all, the tool was planned to support both bioschemas and the original schema we developed with BIII. I would need to revise it a bit, let's see if we can have a look at it soon, maybe during the Elixir BioHackathon.

[matuskalas]

Hi @matuskalas, I checked and it should be fixed now (https://github.com/research-software-ecosystem/utils/pull/33/files).

Looks awesome, many thanks! 🙏🏽

I'm wondering about the BIII URI, please see my question in NEUBIAS/bise#214

For your second question, it's not difficult at all, the tool was planned to support both bioschemas and the original schema we developed with BIII. I would need to revise it a bit, let's see if we can have a look at it soon, maybe during the Elixir BioHackathon.

This will be very nice to have! Especially now that there'll be more activity during the Taggathon in Kobe Sun-Tue.

[albangaignard]

As for now, the bioschemas markup really describes the crawled web pages. As soon as the queried server is the main biii server and not the dev server, the URIs for bioimaging tools will be fixed.

I think we can can close the issue.