Integrate generate_microreact.py script

[johnlees]

The script at scripts/generate_microreact.py should be integrated as an exported command in the main package to avoid duplicating functions.

Additionally, the following should be fixed:

• Use of rapidnj isn't supported (function definition not included)
• --output isn't clear if it is a prefix/directory/file, and fails if a folder with that name does not exist

[johnlees]

The dists.csv file should also be removed

[nickjcroucher]

Added the "--generate-viz" mode to start to address this with update 873f317. This should allow a subset of data to be extracted and visualised as well. I condensed some of the functions within main.py, looks like this may have caused a problem according to the travis checks.

[johnlees]

Great, thanks! Working on sorting out the problems now

[nickjcroucher]

Two issues cropping up with my code. First, there's a problem with the Cytoscape output file:

Building phylogeny Running t-SNE Writing phandango output Copying microreact treeWriting grapetree output Using microreact treeWriting cytoscape output Traceback (most recent call last): File "./PopPUNK/poppunk-runner.py", line 10, in <module> main() File "/lustre/scratch118/infgen/team284/nc3/transmission/PopPUNK/PopPUNK/PopPUNK/__main__.py", line 466, in main outputsForCytoscape(genomeNetwork, isolateClustering, args.output, args.info_csv) File "/lustre/scratch118/infgen/team284/nc3/transmission/PopPUNK/PopPUNK/PopPUNK/plot.py", line 388, in outputsForCytoscape queryList) File "/lustre/scratch118/infgen/team284/nc3/transmission/PopPUNK/PopPUNK/PopPUNK/plot.py", line 534, in writeClusterCsv pd.DataFrame(data=d).to_csv(outfile, columns = colnames, index = False)
Second, there combined clusters being output to the microreact CSV just read "set()". Which appears problematic.

I'll try to have a look at what's going on here.

[nickjcroucher]

To complete the error for Cytoscape above, the final line is:

KeyError: "None of [['id', 'combined_Cluster']] are in the [columns]"

[johnlees]

I should update the travis-CI to do more thorough checks soon, as we've had a few regressions now as features have been added.

First thought on the cytoscape error (might be the same in phandango and grapetree outputs) is that it might be the graph passed needing to be a dict with graph['combined'] = <graph> vs just a

[nickjcroucher]

Microreact, Phandango and Grapetree all completed, albeit with the weird "set()" clustering, it only fell over on the Cytoscape output.

[nickjcroucher]

Fixed in b07ebc2.

[johnlees]

Ah, didn't realise this was specific to the --generate-viz mode. Obviously I didn't check what the readClusters() function actually did! Surprised it got that far at all. Thanks for fixing!