In [2]:

    

import numpy as np 
import matplotlib as mpl 
import pandas as pd
output = pd.read_csv('./data/Gradients.csv')
output.head()


    

    
Out[2]:






  

    

      

      
SUB_ID
      
X
      
subject
      
SITE_ID
      
FILE_ID
      
DX_GROUP
      
DSM_IV_TR
      
AGE_AT_SCAN
      
SEX
      
HANDEDNESS_CATEGORY
      
...
      
1
      
2
      
3
      
4
      
5
      
6
      
7
      
8
      
9
      
file
    
  
  

    

      
0
      
50004
      
3
      
50004
      
PITT
      
Pitt_0050004
      
1
      
1
      
19.09
      
1
      
R
      
...
      
0.008470
      
0.007940
      
0.007781
      
0.008144
      
0.009179
      
0.008527
      
0.008443
      
0.007053
      
0.007298
      
Pitt_0050004_rois_cc400.1D.npy
    
    

      
1
      
50009
      
8
      
50009
      
PITT
      
Pitt_0050009
      
1
      
1
      
33.86
      
1
      
R
      
...
      
0.008657
      
0.008265
      
0.008871
      
0.008278
      
0.008208
      
0.007907
      
0.007276
      
0.008844
      
0.008266
      
Pitt_0050009_rois_cc400.1D.npy
    
    

      
2
      
50010
      
9
      
50010
      
PITT
      
Pitt_0050010
      
1
      
1
      
35.20
      
1
      
L
      
...
      
0.008397
      
0.007800
      
0.007688
      
0.006135
      
0.007923
      
0.007170
      
0.007871
      
0.006554
      
0.005794
      
Pitt_0050010_rois_cc400.1D.npy
    
    

      
3
      
50012
      
11
      
50012
      
PITT
      
Pitt_0050012
      
1
      
1
      
21.48
      
1
      
R
      
...
      
0.008173
      
0.008331
      
0.008421
      
0.008395
      
0.007317
      
0.008295
      
0.007617
      
0.007579
      
0.008206
      
Pitt_0050012_rois_cc400.1D.npy
    
    

      
4
      
50020
      
18
      
50020
      
PITT
      
Pitt_0050020
      
1
      
1
      
20.83
      
1
      
R
      
...
      
0.008207
      
0.008663
      
0.007796
      
0.008236
      
0.007974
      
0.008120
      
0.008483
      
0.010089
      
0.007470
      
Pitt_0050020_rois_cc400.1D.npy
    
  

5 rows × 118 columns

In [20]:

    

# Column names are strings, so need to provide a list of strings to index:
df = output[['DX_GROUP'] + [str(x) for x in range(10)]]
ASC = df['DX_GROUP'] == 2
NT = df['DX_GROUP'] == 1
G1 = df[ASC]
G2 = df[NT]


    

In [22]:

    

%matplotlib inline

## agg backend is used to create plot as a .png file
mpl.use('agg')

import matplotlib.pyplot as plt 

# some plotting options
fs = 10  # fontsize
flierprops = dict(marker='o', markerfacecolor='green', markersize=12,
                  linestyle='none')

## combine the group collections into a list    
Grd0 = [G1['0'], G2['0']]
Grd1 = [G1['1'], G2['1']]
Grd2 = [G1['2'], G2['2']]
Grd3 = [G1['3'], G2['3']]
Grd4 = [G1['4'], G2['4']]
Grd5 = [G1['5'], G2['5']]
Grd6 = [G1['6'], G2['6']]
Grd7 = [G1['7'], G2['7']]
Grd8 = [G1['8'], G2['8']]
Grd9 = [G1['9'], G2['9']]

fig, axes = plt.subplots(nrows=2, ncols=5, figsize=(6, 6), sharey=True)

axes[0, 0].boxplot(Grd0, patch_artist=True)
axes[0, 0].set_title('G0', fontsize=fs)

axes[0, 1].boxplot(Grd1, patch_artist=True)
axes[0, 1].set_title('G1', fontsize=fs)

axes[0, 2].boxplot(Grd2, patch_artist=True)
axes[0, 2].set_title('G2', fontsize=fs)

axes[0, 3].boxplot(Grd3, patch_artist=True)
axes[0, 3].set_title('G3', fontsize=fs)

axes[0, 4].boxplot(Grd4, patch_artist=True)
axes[0, 4].set_title('G4', fontsize=fs)

axes[1, 0].boxplot(Grd5, patch_artist=True)
axes[1, 0].set_title('G5', fontsize=fs)

axes[1, 1].boxplot(Grd6, patch_artist=True)
axes[1, 1].set_title('G6', fontsize=fs)

axes[1, 2].boxplot(Grd7, patch_artist=True)
axes[1, 2].set_title('G7', fontsize=fs)

axes[1, 3].boxplot(Grd8, patch_artist=True)
axes[1, 3].set_title('G8', fontsize=fs)

axes[1, 4].boxplot(Grd9, patch_artist=True)
axes[1, 4].set_title('G9', fontsize=fs)

fig.suptitle("Gradient Slopes")
fig.subplots_adjust(hspace=0.4)