Sort Richness Plots by Observed count #343
Conversation
add an argument and the code to allow sorting by observed counts in the richness plots
|
Example plot on my data with and without sorting:
|
|
Thanks! I love pull requests. I'll check it out. |
| }else{ | ||
| # Make the ggplot. | ||
| p <- ggplot(mdf, richness_map) + geom_point(na.rm=TRUE) | ||
| } |
There was a problem hiding this comment.
I think I'd prefer this to be a more general sortby argument, where the sortby can be any of the measures or sample_variables. There is analogous sorting support in the plot_heatmap function for sorting either the OTU or sample indices (or both) by a specified variable. In this way, "Observed" would just be one possible argument of many. If more general I think this argument would be worth supporting and I would merge.
Thanks! And let me know if you're willing to mock-up the more general case. I don't want to merge as-is because I don't want to temporarily add an argument I'm soon going to replace with something else...
|
Hi Joey711, thanks for taking a look. I'll try porting to a general version using zach cp |
Use a generic sortby that can sort using sample_data or measurements. Add a unit test to test each of those.
|
HI Joey711, Sorry it took me a while to get to this but I've put together a simple function that can sort on any column in the master DF (sample_data and richness measurement). I put a simple unittest and example parameter but am happy to change those. I can also sqush the commits into a single commit to make the history cleaner. thanks again for your patience and your great software package. |
|
Thanks @zachcp ! I'll need to take a look, and run tests, but don't worry abut condensing the commits. I'll most likely pull this into a separate branch, and add any necessary tests/changes on top. Thanks for the contribution! |
|
Yep, I'm working on it. See https://github.com/joey711/phyloseq/tree/pr/343 |
As a followup to #342 I put together a working solution. I don't know if/how you would want this feature but I put together a working version for you to look at.