Visualisation of Protein Graphs

[a-r-j]

• [x] 3d Interactive Viewer (using coords) - plotly
• [x] 3d static plot (using coords) - Nx & Matplotlib
• [x] 2d projection

Colouring options

• [x] Resiude type
• [x] Atom type
• [x] Edge types
• [x] Degree
• [x] Sequence position

Node Size options

• [x] Degree
• [ ] By feature magnitude (single or several?)

Labelling options

• [x] Residue Name
• [x] Sequence Position
• [x] Identifier

Documentation

• [x] Docstrings
• [x] Docs

[kamurani]

Hi @a-r-j

Do we still want scaling node size by By feature magnitude (single or several?) as a feature?

I'm currently implementing it for a project anyway, but would be happy to submit a PR if we want an inbuilt way of doing so. maybe passing in a node_size_feature to plot_protein_structure_graph().

What do you reckon?

[a-r-j]

Hi @cimranm that'd be a great addition.

I'm not sure as to how to handle it most elegantly. One option is to use a string argument to grab a node feature from the graph (easier for the user maybe if only using scalar features, but a lot of them are stored as series/arrays etc) or node_size_feature is a dict

node_size_feature: Dict[str, float] = {"A:SER:2": 1.2, "A:GLY:3": 3.4, ...}

or an array of length num nodes:

node_size_feature: np.ndarray = np.array([1.2, 3.4,...])
len(node_size_feature) == len(g)

What do you think?

[kamurani]

Hmm, I've had a chance to look at it and i've given another PR. Not sure if the way i'm implementing it makes sense (I'm still new to this!) Would love to hear your thoughts @a-r-j