[BUG] secondary_structure_percentage doctest raises pandas-related AttributeError

[FloLangenfeld]

I just launch the doctest code written in the rstoolbox.analysis.secondary_structure_percentage function docstring:

from rstoolbox.io import parse_rosetta_file
from rstoolbox.analysis import secondary_structure_percentage
import pandas as pd
pd.set_option('display.width', 1000)
pd.set_option('display.max_columns', 500)
df = parse_rosetta_file("../rstoolbox/tests/data/input_ssebig.minisilent.gz",
                        {'scores': ['score'], 'structure': 'C'})
df = secondary_structure_percentage(df, 'C')
df.head()

It raises the following error:

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/florent/anaconda3/envs/rstoolbox/lib/python3.8/site-packages/rstoolbox/analysis/structure.py", line 240, in secondary_structure_percentage
    df2 = df.apply(lambda row: secondary_structure_percentage(row,
  File "/home/florent/anaconda3/envs/rstoolbox/lib/python3.8/site-packages/pandas/core/frame.py", line 7541, in apply
    return op.get_result()
  File "/home/florent/anaconda3/envs/rstoolbox/lib/python3.8/site-packages/pandas/core/apply.py", line 180, in get_result
    return self.apply_standard()
  File "/home/florent/anaconda3/envs/rstoolbox/lib/python3.8/site-packages/pandas/core/apply.py", line 255, in apply_standard
    results, res_index = self.apply_series_generator()
  File "/home/florent/anaconda3/envs/rstoolbox/lib/python3.8/site-packages/pandas/core/apply.py", line 284, in apply_series_generator
    results[i] = self.f(v)
  File "/home/florent/anaconda3/envs/rstoolbox/lib/python3.8/site-packages/rstoolbox/analysis/structure.py", line 240, in <lambda>
    df2 = df.apply(lambda row: secondary_structure_percentage(row,
  File "/home/florent/anaconda3/envs/rstoolbox/lib/python3.8/site-packages/rstoolbox/analysis/structure.py", line 253, in secondary_structure_percentage
    dfp = df.append(pd.Series([float(csse['H']) / len(sse), float(csse['E']) / len(sse),
  File "/home/florent/anaconda3/envs/rstoolbox/lib/python3.8/site-packages/pandas/core/series.py", line 2688, in append
    return concat(
  File "/home/florent/anaconda3/envs/rstoolbox/lib/python3.8/site-packages/pandas/core/reshape/concat.py", line 274, in concat
    op = _Concatenator(
  File "/home/florent/anaconda3/envs/rstoolbox/lib/python3.8/site-packages/pandas/core/reshape/concat.py", line 395, in __init__
    axis = sample._constructor_expanddim._get_axis_number(axis)
AttributeError: 'function' object has no attribute '_get_axis_number'

I did a quick test with pandas 0.24.1 and it worked fine, so my best guess is that there was a change in the _Concatenator class for the pandas v1 major update.

I kind of solved that issue with a quick and dirty hack in rstoolbox.analysis.structure.py (just tested in my pandas 1.1.0 python environment). So if your interested, just let me know and I'll submit a PR.

>>> rstoolbox.show_versions():
yaml: 5.3.1
pandas: 1.1.0
seaborn: 0.10.1
libconfig: 0.9.2
six: 1.15.0
networkx: 2.4
>>> rstoolbox.__version__
'1.0.1'

[jaumebonet]

Hi @FloLangenfeld !
Thanks for your interest!
This comes, of course, from the bigger issue, which is the general integration of pandas 1 into the code 😕

I also need to fix the CI with travis, as it has changed a bit since last time I checked this code, but if you open the PR I'll be happy to merge it as soon as I can.

[FloLangenfeld]

@jaumebonet
I've opened the PR with the small trick here : #124

And yeah, pandas 1 was a game-changer in so many ways that one may ask why not just add pandas<1.0.0 in the requirements 😆