Output experimental data paths for use in F/E

[mwinokan]

To shorten the feedback loop between alignment, upload, and reviewing of data in case there are faulty observations, @phraenquex suggests that we need some UI to be able to see the paths to the original experimental data for each observation.

This will require collator/aligner changes in XCA @tdudgeon

F/E UI specs still to follow @matej-vavrek

[phraenquex]

What is needed is the frontend to know the original location of the PDB file - the thing that's in soakDB.

@tdudgeon needs to add to XCA/yamls
@kaliif needs to load and serve by API
@matej-vavrek needs to stick it into the extended view (probably a [copy] button next to the longcode)

[tdudgeon]

Collator now adds source_file property to the crystallographic_files entries.
Example:

    crystallographic_files:
      xtal_pdb:
        file: upload_1/crystallographic_files/A71EV2A-x0450/A71EV2A-x0450.pdb
        sha256: dec53789027de5ffc0eb42b4158c1023f96d59142291e1b67676eeaef4a17df2
        source_file: data/lb32627-66/inputs/dls/labxchem/data/lb32627/lb32627-66/processing/analysis/model_building/A71EV2A-x0450/Refine_0004/refine_4.split.bound-state.pdb
      xtal_mtz:
        file: upload_1/crystallographic_files/A71EV2A-x0450/A71EV2A-x0450.mtz
        sha256: 48ce6ce332b71246126912d05ce3ac1ab1452b9203c382c279c2fb933c4ca178
        source_file: data/lb32627-66/inputs/dls/labxchem/data/lb32627/lb32627-66/processing/analysis/model_building/A71EV2A-x0450/Refine_0004/refine_4.mtz
      ligand_cif:
        file: upload_1/crystallographic_files/A71EV2A-x0450/A71EV2A-x0450.cif
        sha256: d55dadae7fb0d7638b249b27d03c1f3207ab6065527fd6ca2d067c63759a9ee1
        source_file: data/lb32627-66/inputs/dls/labxchem/data/lb32627/lb32627-66/processing/analysis/model_building/A71EV2A-x0450/compound/Z100642432.cif
        smiles: CN(C)C(=O)c1ccc(F)cc1Br
        ligand_name: LIG
      ligand_binding_events:
      - file: upload_1/crystallographic_files/A71EV2A-x0450/A71EV2A-x0450_1_A_201.ccp4
        sha256: 56d493d405bd9bfa8105b7bba96adb3741bdac7ea13d96293fcec6f8bc59a780
        source_file: data/lb32627-66/inputs/dls/labxchem/data/lb32627/lb32627-66/processing/analysis/pandda2_remaining_screen_smallcell_2023-12-05/processed_datasets/A71EV2A-x0450/A71EV2A-x0450-event_2_1-BDC_0.9_map.native.ccp4
        model: '1'
        chain: A
        res: 201
        index: 2
        bdc: 0.9

Commit: xchem/xchem-align@c6e6ded

Example data: https://im-xchem-data.s3.amazonaws.com/lb32627-66_18-SEP.tgz

[kaliif]

@matej-vavrek source paths are available in api/experiments/. Site observation objects have a key to experiment.

[phraenquex]

@tdudgeon, @kaliif that's fantastic!

One detail: ensure it's the absolute path (which it currently isn't).

[mwinokan]

@tdudgeon to change to absolute paths (and then update the environment together with #1511)

[mwinokan]

@matej-vavrek please add a clipboard icon to the F/E observation expanded table under a heading "Path"

[mwinokan]

Latest alignment from A71EV2A on staging includes the absolute paths in the /api/experiments endpoint

[matej-vavrek]

@phraenquex, @mwinokan, I have updated my stack. Currently, it uses pdb_info for the path from api/experiments.

[mwinokan]

Thanks @matej-vavrek the correct field is pdb_info_source_file which will be populated in new alignments

[matej-vavrek]

@phraenquex, @mwinokan, changed field and also added a note into tooltip when it is empty, is it ok?

[mwinokan]

@matej-vavrek please push this to staging so we can test with new data

[phraenquex]

@mwinokan @matej-vavrek should I be expecting to see this in staging for the A71EV2A you uploaded yesterday? Because I don't....

https://fragalysis.xchem.diamond.ac.uk/viewer/react/projects/100/71

[mwinokan]

@phraenquex it was briefly there but staging is out of date (see slack thread) they should reappear on the next staging rebuild

[phraenquex]

@mwinokan is this the path we'd expect - specifically the CIF file?

[phraenquex]

@kaliif the path that should be served to the frontend is the path to the PDB file.

Fix that, and then we can close the ticket.

[mwinokan]

@matej-vavrek says that the f/e is using querying:

experiment.pdb_info_source_file

@kaliif it seems that you are currently serving the source_file value from the ligand_cif key. but it should be the source_file valuextal_pdb

[mwinokan]

F/E is now correctly serving the refined PDB, e.g. for A71EV2A A0207a:

/dls/labxchem/data/lb32627/lb32627-66/processing/analysis/model_building/A71EV2A-x0207/Refine_0003/refine_3.split.bound-state.pdb