Graph plots #197
Graph plots #197
Conversation
|
Testing asteroid plot, scaling by RSA and colour by hydrophobicity fig1 = asteroid_plot(
g=g,
node_id=node,
colour_nodes_by="hydrophobicity",
size_nodes_by='rsa',
node_size_multiplier=80
)
|
|
@a-r-j any changes you want me to make? |
Hey, sorry - busy period. A few minor comments but otherwise LGTM. Thanks for the contributions! |
graphein/protein/visualisation.py
Outdated
| return hmap[res] | ||
|
|
||
| node_colours = [] | ||
| for n in subgraph.nodes(): |
There was a problem hiding this comment.
I believe you can replace this nested loop with:
for n, d in subgraph.nodes(data=True)
...…ith more features
…ding which scale to use; or just return vector of values for all 5 scales.
| TODO: Functino that gets ``min`` and ``max`` values in a graph for a given feature so that we can scale / offset to > 0 | ||
| TODO: should feature `distance` actually contain the site itself i.e. in the string? | ||
| """ | ||
| def _node_feature_func( |
There was a problem hiding this comment.
This is really elegant work - thanks a lot!
Having taken a closer look at the code I wonder if we can be even smarter here. Given the rest of Graphein is quite functional in style, we could perhaps refactor out the sizing functions into atomic functions & then allow users to pass in either a function of their own or a string which would dispatch one the lambdas/more complicated functions you've implemented here. I think this provides the most flexibility whilst supporting all the great features you've added here. What do you think?
|
SonarCloud Quality Gate failed.
|
|
SonarCloud Quality Gate failed.
|
|
Kudos, SonarCloud Quality Gate passed!
|
|
|
2 participants
Reference Issues/PRs
couldn't find an issue that references this
What does this implement/fix? Explain your changes
add support for colouring / sizing asteroid plot nodes by different features. Extra functions can be added for more features if it is known how to reference the node data e.g.
g.nodes(data=True)[k]['rsa']change colourscheme to viridis (can revert this if it's unwanted)
What testing did you do to verify the changes in this PR?
Plotted example graphs.
Pull Request Checklist
./CHANGELOG.mdfile (if applicable)./graphein/tests/*directories (if applicable)./notebooks/(if applicable)python -m py.test tests/and make sure that all unit tests pass (for small modifications, it might be sufficient to only run the specific test file, e.g.,python -m py.test tests/protein/test_graphs.py)black .andisort .