testing and fixing outdated atlas generation scripts

[viktorpm]

Description

We want to test if the atlas validation functions pass after regenerating atlases (see: #201).
Currently, the atlas generation scripts are outdated, so to regenerate the atlases, first, we need to fix and tidy the scripts.

What is this PR

• Bug fix
• Addition of a new feature
• Other

Why is this PR needed?

We need an up-to-date version of the atlas generation scripts and the atlases.

What does this PR do?

Fixes syntax errors, and removes outdated packages.

References

#201

How has this PR been tested?

It's just a draft PR so far

Is this a breaking change?

It might be, due to changes in vedo syntax and function arguments.

Does this PR require an update to the documentation?

Yes

Checklist:

• The code has been tested locally
• Tests have been added to cover all new functionality (unit & integration)
• The documentation has been updated to reflect any changes
• The code has been formatted with pre-commit

[viktorpm]

Solved errors:

Script - error number	Error	Note	Solved
admba_3d_dev_mouse.py - 1	Mesh.decimate() got an unexpected keyword argument 'method'	replace with decimate_pro()	Yes, tested locally & on HPC, commit: 0c1584a
allen_cord.py - 1	BadZipFile: File is not a zip file	fixing source URL	Yes, tested locally, commit: 5d4f4b0
allen_mouse.py	no error	-	-
azba_zfish.py - 1	AttributeError: 'Mesh' object has no attribute 'smoothLaplacian'	replace smoothLaplacian with smooth in mesh_utils.py	Yes, tested on HPC, commit: a8f4631
example_mouse.py	no error	-	-
kim_developmental_ccf_mouse.py - 1	AttributeError: 'DataFrame' object has no attribute 'append'	possible pandas issue	Yes, tested on HPC, commit: 58495b6
kim_mouse.py - 1	Mesh.decimate() got an unexpected keyword argument 'method'	replace with decimate_pro()	Yes, tested on HPC, creating meshes is very slow!
mpin_zfish.py - 1	allensdk	pip install brainglobe-atlasapi[allenmouse]	Yes
osten_mouse.py -1	allensdk	pip install brainglobe-atlasapi[allenmouse]	Yes, tested locally & on HPC
princeton_mouse.py	no error	-	-
whs_sd_rat.py	no error	-	-

Errors to be solved:

Script - error number	Error	Note	Solved
admba_3d_dev_mouse.py - 2	gets stuck at region 112892417 of admba_3d_p28_mouse	Creating mesh for region 112892417 No labels found for r10LimCG (112892417)	no
admba_3d_dev_mouse.py - 3	doesn't skip already generated atlases	FileExistsError: [Errno 17] File exists:	no
humanatlas.py - 1	FileNotFoundError: No such file or no access: 'D:\Dropbox (UCL -SWC...'	hardcoded paths in the code	no, needs separate PR
mpin_zfish.py - 2	SSLError: HTTPSConnectionPool(...)	-	no
perens_lsfm_mouse.py - 1	IndexError: tuple index out of range		no

[viktorpm]

Steps to test changes on the HPC:

module load mamba
mamba init
mamba create -n bg-atlasapi python=3.11
mamba activate bg-atlasapi
git clone https://github.com/brainglobe/brainglobe-atlasapi.git
cd brainglobe-atlasapi/
git checkout fixing_atlas_scripts
pip install -e ".[allenmouse, dev]"
pip install scikit-image
pip install imio
pip install brainio
cd brainglobe_atlasapi/atlas_generation/atlas_scripts

[alessandrofelder]

Looking good - think we just need to fix the dependencies in pyproject.toml and then this is good to go 🚀

[alessandrofelder]

This should be added to the dependencies in pyproject.toml then, I think,

[alessandrofelder]

Again, worth adding to the dependencies. If it is very big or hard-to-install, it can be an optional dependency like allensdk.

[alessandrofelder]

We should depend on brainglobe-utils.imageio instead (see brainglobe/brainglobe-utils#33 for details), which may need to be added to the dependencies.