G unifrac #19
G unifrac #19
Conversation
|
@wasade Can you update this branch with the latest version of master? |
|
Can you update the branch again :\
|
|
BAM BAM |
q2_state_unifrac/plugin_setup.py
Outdated
| 'the table.') | ||
| }, | ||
| outputs=[('distance_matrix', DistanceMatrix % Properties('phylogenetic'))], | ||
| name='Wweighted normalized UniFrac', |
There was a problem hiding this comment.
Wweighted -> Weighted 🏋️ 🏋️♀️
| description=('This method computes weighted unnormalized UniFrac') | ||
| description=('This method computes weighted unnormalized UniFrac as ' | ||
| 'described in Lozupone et al. 2007 Appl Environ Microbiol; ' | ||
| 'DOI: 10.1128/AEM.01996-06') |
There was a problem hiding this comment.
Would be nice to add the citation_text for all these papers, see for example:
There was a problem hiding this comment.
Thanks! Isn't that at the plugin level though, and not at the method level? These citations are method specific
sucpp/test_su.cpp
Outdated
| w_task_p.g_unifrac_alpha = 1.0; | ||
| su::unifrac(table, tree, su::generalized, w_strides, w_strides_total, &w_task_p); | ||
|
|
||
| // as computed by GUniFrac |
There was a problem hiding this comment.
For sanity in the future, might be a good idea to include package version that produced this output.
sucpp/test_su.cpp
Outdated
| d0_task_p.g_unifrac_alpha = 0.0; | ||
| su::unifrac(table, tree, su::generalized, d0_strides, d0_strides_total, &d0_task_p); | ||
|
|
||
| // as computed by GUniFrac |
| const su::task_parameters* task_p) { | ||
|
|
||
| if(table.n_samples != task_p->n_samples) { | ||
| fprintf(stderr, "Task and table n_samples not equal\n"); |
There was a problem hiding this comment.
table is not passed into the kernels, so table->n_samples is not accessible without a lot of monkeying. This is just a sanity check that the task parameters reflect the table (loosely)
sucpp/unifrac.hpp
Outdated
| unsigned int end, | ||
| unsigned int tid); | ||
| const task_parameters* task_p); | ||
| //void unweighted_unifrac(biom &table, |
There was a problem hiding this comment.
Can you remove these lines?
Depends on #14 and #18, merge those first
We currently validate
a=1.0. TheGUniFracR package does not include tests unfortunately. Before merge, it would be good to include verification ofa=0.0anda=0.5. Note thata=0.0is unweighted UniFrac only if the proportions are dichotomized.