phenotype data frame format for association plot

[alejandrocs98]

Hi! I am trying to use micom.viz.association.plot_association() but it is not very clear for me how the input for phenotype argument should be. I am using a pandas DataFrame with the index coinciding with the values of sample_id in the GrowtthResults object and one column called phenotype that contain one of two values (as I understood should be from the doc):
minimal_imports

GrowthResults(growth_rates=     abundance   growth_rate  reactions  metabolites                   taxon  \
0     0.097084  1.231692e-14       2381         1461                B_caccae   
1     0.444060  3.826183e-02       2779         1702  B_cellulosilyticus_WH2   
2     0.139322  1.136146e-22       2476         1526                B_ovatus   
3     0.093460  1.196529e-01       2524         1549      B_thetaiotaomicron   
4     0.035084  1.535031e-22       2418         1500             B_uniformis   
..         ...           ...        ...          ...                     ...   
160   0.004272  1.445010e-06        920          850           C_aerofaciens   
161   0.019892  4.351062e-07       2049         1394              C_scindens   
162   0.000125  2.261928e-04       1980         1313            C_spiroforme   
163   0.034353  9.585099e-03       2519         1570            P_distasonis   
164   0.009636  5.057161e-06       1225         1088                 R_obeum   

     tradeoff sample_id  
0         0.6  mouse_01  
1         0.6  mouse_01  
2         0.6  mouse_01  
3         0.6  mouse_01  
4         0.6  mouse_01  
..        ...       ...  
160       0.6  mouse_15  
161       0.6  mouse_15  
162       0.6  mouse_15  
163       0.6  mouse_15  
164       0.6  mouse_15  
...
1402        C00378            1130.0       EX_thm_m  
1403        C05529            1084.0      EX_tsul_m  

[1404 rows x 10 columns])

phenotype

          diet_type
sample_id          
mouse_01     LF/HPP
mouse_02     LF/HPP
mouse_03     LF/HPP
mouse_04     LF/HPP
mouse_05     LF/HPP
mouse_06     LF/HPP
mouse_07     LF/HPP
mouse_08      HF/HS
mouse_09      HF/HS
mouse_10      HF/HS
mouse_11      HF/HS
mouse_12      HF/HS
mouse_13      HF/HS
mouse_14      HF/HS
mouse_15      HF/HS

from micom.viz import association
association.plot_association(
    minimal_imports, 
    phenotype, 
    'binary', 
    'diet type', 
    'results/associations/associations.html'
    )

I am getting the following error:

alueError                                Traceback (most recent call last)
Cell In[14], [line 1](vscode-notebook-cell:?execution_count=14&line=1)
----> [1](vscode-notebook-cell:?execution_count=14&line=1) association.plot_association(
      [2](vscode-notebook-cell:?execution_count=14&line=2)     minimal_imports, 
      [3](vscode-notebook-cell:?execution_count=14&line=3)     phenotype, 
      [4](vscode-notebook-cell:?execution_count=14&line=4)     'binary', 
      [5](vscode-notebook-cell:?execution_count=14&line=5)     'diet_type', 
      [6](vscode-notebook-cell:?execution_count=14&line=6)     'results/associations/associations.html'
      [7](vscode-notebook-cell:?execution_count=14&line=7)     )

File [~/miniconda3/envs/micom-test/lib/python3.9/site-packages/micom/viz/association.py:87](https://file+.vscode-resource.vscode-cdn.net/home/acs98/Documents/BCEM/Tesis-maestria/masters-thesis/metabolic-modeling/micom/~/miniconda3/envs/micom-test/lib/python3.9/site-packages/micom/viz/association.py:87), in plot_association(results, phenotype, variable_type, variable_name, filename, flux_type, fdr_threshold, threads, atol)
     [85](https://file+.vscode-resource.vscode-cdn.net/home/acs98/Documents/BCEM/Tesis-maestria/masters-thesis/metabolic-modeling/micom/~/miniconda3/envs/micom-test/lib/python3.9/site-packages/micom/viz/association.py:85) if exchanges.shape[1] < 1:
     [86](https://file+.vscode-resource.vscode-cdn.net/home/acs98/Documents/BCEM/Tesis-maestria/masters-thesis/metabolic-modeling/micom/~/miniconda3/envs/micom-test/lib/python3.9/site-packages/micom/viz/association.py:86)     raise ValueError("None of the fluxes passed the tolerance threshold :(")
---> [87](https://file+.vscode-resource.vscode-cdn.net/home/acs98/Documents/BCEM/Tesis-maestria/masters-thesis/metabolic-modeling/micom/~/miniconda3/envs/micom-test/lib/python3.9/site-packages/micom/viz/association.py:87) if variable_type == "binary" and phenotype.nunique() != 2:
     [88](https://file+.vscode-resource.vscode-cdn.net/home/acs98/Documents/BCEM/Tesis-maestria/masters-thesis/metabolic-modeling/micom/~/miniconda3/envs/micom-test/lib/python3.9/site-packages/micom/viz/association.py:88)     raise ValueError(
     [89](https://file+.vscode-resource.vscode-cdn.net/home/acs98/Documents/BCEM/Tesis-maestria/masters-thesis/metabolic-modeling/micom/~/miniconda3/envs/micom-test/lib/python3.9/site-packages/micom/viz/association.py:89)         "Binary variables must have exactly two unique values, yours "
     [90](https://file+.vscode-resource.vscode-cdn.net/home/acs98/Documents/BCEM/Tesis-maestria/masters-thesis/metabolic-modeling/micom/~/miniconda3/envs/micom-test/lib/python3.9/site-packages/micom/viz/association.py:90)         "has: %s." % ", ".join(phenotype.unique())
     [91](https://file+.vscode-resource.vscode-cdn.net/home/acs98/Documents/BCEM/Tesis-maestria/masters-thesis/metabolic-modeling/micom/~/miniconda3/envs/micom-test/lib/python3.9/site-packages/micom/viz/association.py:91)     )
     [92](https://file+.vscode-resource.vscode-cdn.net/home/acs98/Documents/BCEM/Tesis-maestria/masters-thesis/metabolic-modeling/micom/~/miniconda3/envs/micom-test/lib/python3.9/site-packages/micom/viz/association.py:92) elif variable_type == "continuous" and not is_numeric_dtype(phenotype):

File [~/miniconda3/envs/micom-test/lib/python3.9/site-packages/pandas/core/generic.py:1576](https://file+.vscode-resource.vscode-cdn.net/home/acs98/Documents/BCEM/Tesis-maestria/masters-thesis/metabolic-modeling/micom/~/miniconda3/envs/micom-test/lib/python3.9/site-packages/pandas/core/generic.py:1576), in NDFrame.__nonzero__(self)
   [1574](https://file+.vscode-resource.vscode-cdn.net/home/acs98/Documents/BCEM/Tesis-maestria/masters-thesis/metabolic-modeling/micom/~/miniconda3/envs/micom-test/lib/python3.9/site-packages/pandas/core/generic.py:1574) @final
   [1575](https://file+.vscode-resource.vscode-cdn.net/home/acs98/Documents/BCEM/Tesis-maestria/masters-thesis/metabolic-modeling/micom/~/miniconda3/envs/micom-test/lib/python3.9/site-packages/pandas/core/generic.py:1575) def __nonzero__(self) -> NoReturn:
-> [1576](https://file+.vscode-resource.vscode-cdn.net/home/acs98/Documents/BCEM/Tesis-maestria/masters-thesis/metabolic-modeling/micom/~/miniconda3/envs/micom-test/lib/python3.9/site-packages/pandas/core/generic.py:1576)     raise ValueError(
   [1577](https://file+.vscode-resource.vscode-cdn.net/home/acs98/Documents/BCEM/Tesis-maestria/masters-thesis/metabolic-modeling/micom/~/miniconda3/envs/micom-test/lib/python3.9/site-packages/pandas/core/generic.py:1577)         f"The truth value of a {type(self).__name__} is ambiguous. "
   [1578](https://file+.vscode-resource.vscode-cdn.net/home/acs98/Documents/BCEM/Tesis-maestria/masters-thesis/metabolic-modeling/micom/~/miniconda3/envs/micom-test/lib/python3.9/site-packages/pandas/core/generic.py:1578)         "Use a.empty, a.bool(), a.item(), a.any() or a.all()."
   [1579](https://file+.vscode-resource.vscode-cdn.net/home/acs98/Documents/BCEM/Tesis-maestria/masters-thesis/metabolic-modeling/micom/~/miniconda3/envs/micom-test/lib/python3.9/site-packages/pandas/core/generic.py:1579)     )

ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

Which might be a bug or a problem related to my python version (I believe that just using phenotype.nunique().iloc[0] != 2 or something similar should fix this issue), but even though I am not sure if I am giving the proper inputs to the function.
Thanks :)

[cdiener]

Hi, sorry about being unclear. One is supposed to pass the column itself, so phenotype.diet_type in your case. The name you pass in is just the legend you want in the visualization (does not have to be the column name).

I will add a type check to the function.