191 validate cross links based on armstrong error tolerance

[AnnaPolensky]

Description

fixes #191
There is a new step where the intra-crosslinks of one protein are being validated (-> the user can enter an allowed deviation from the crosslinker's length. If the distance of the connected amino acids in alphafold is greater than crosslinker length + allowed deviation, the crosslinker will be marked as invalid. The user can not only apply an upper bound but also a lower bound to the allowed deviation.)
Right now, the distance calculation is based on the central Ca atom; we might want to change this to the residue atom where the crosslinker actually binds later on.
The output is a simple bar plot and we add two colums to the crosslinking df (distance in alphafold and true or false on whether the crosslink matches the alphafold data or not).

Changes

All the functions for the actual validation are in backend/protzilla/data_analysis/cross_linking_validation.py
The form is defined at the end of backend/protzilla/methods/data_analysis.py
Simple tests are in backend/tests/protzilla/data_analysis/test_crosslinking_validation.py

Testing

1. Add Crosslinking Data Import and import p26splussubstrate_XL.xlsx
2. Add Alphafold Load Step and load O43242
3. Add Crosslinking Validation (this is a data analysis step), select O43242, DSSO should have a length of 10.2. Test different values for the bounds. E.g. if we only apply an upper bound of 1, only one crosslink should match the Alphafold data, for 2, 2 crosslinks should match.

Check that in the crosslinking validation step three fields show up for each crosslinker that appears in the crosslinking data (for p26splussubstrate_XL.xlsx only DSSO should show up).

PR checklist

Development

• If necessary, I have updated the documentation (README, docstrings, etc.)
• If necessary, I have created / updated tests.

Mergeability

• main-branch has been merged into local branch to resolve conflicts
• The tests and linter have passed AFTER local merge
• The backend code has been formatted with black
• The frontend code has been formatted with pnpm format and checked with pnpm lint

Code review

• I have self-reviewed my code.
• At least one other developer reviewed and approved the changes

[github-actions]

Coverage report

Click to see where and how coverage changed

File	Statements	Missing	Coverage	Coverage (new stmts)	Lines missing
backend/main
views.py					48-59
backend/protzilla
form.py
networking.py					27-32
runner.py
stepfactory.py
steps.py
backend/protzilla/data_analysis
crosslinking_validation.py					55, 120, 210-211, 262-271
differential_expression_anova.py
differential_expression_kruskal_wallis.py
differential_expression_linear_model.py
differential_expression_mann_whitney.py
differential_expression_t_test.py
backend/protzilla/data_integration
di_plots.py
backend/protzilla/importing
alphafold_protein_structure_load.py					68, 80, 124-125, 133-137, 141-142, 161-162, 173-176, 180, 213-214, 219, 223
crosslinking_import.py					105-123, 160-163, 231-254, 277-287, 307-354, 375-421, 472, 474, 477-479, 567-602, 665, 683-691
import_utils.py
ms_data_import.py
backend/protzilla/methods
data_analysis.py					2478, 2521-2538
data_integration.py
importing.py					427, 453
backend/protzilla/utilities
utilities.py
Project Total

This report was generated by python-coverage-comment-action

[Elena-kal]

Seems fine overall. The test as you described worked. Although I think because you fetch the alphafold data in your code yourself one does not need to add the step to upload it. I think in future this should not be included in the code but I am fine with leaving it as it is for now, if you open an issue for it as soon as this is merged.

[Elena-kal]

I was just wondering: How do we distinguish between the same kinds of atoms at different positions? Say we have another C atom, how do we know which one is the central?

[AnnaPolensky]

For each amino acid there is exactly one atom called "CA" (for C alpha). This is the central C atom. The other C atoms are named differently.

[Elena-kal]

Okay I see, what about other atoms like Hydrogen? There are probably several of them in one amino acid right? Do they also have definite naming? No need to change anything here but I 'd be interested to know. Maybe we can talk about it on Tuesday.

[AnnaPolensky]

I'm unsure, I would talk to Chris about this on Tuesday.

[jorisfu]

Just a little thing but looks very good