B. classification¶

Author: Shanshan Yang, Bioinformatics Core, ASU Time: Dec 11, 2017

import pandas as pd
import numpy as np
%matplotlib inline
import matplotlib.pyplot as plt
import seaborn as sns

datafile="B.classification.csv"
data=pd.read_csv(datafile)
data.head()

onehot encoding for categorical variables¶

# Create a new variable for every unique value of TP53mstatus
mstatus = pd.get_dummies( data.mstatus , prefix='mstatus' )
mstatus

mstatus.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 17 entries, 0 to 16
Data columns (total 3 columns):
mstatus_0    17 non-null uint8
mstatus_1    17 non-null uint8
mstatus_2    17 non-null uint8
dtypes: uint8(3)
memory usage: 123.0 bytes

mstatus['mstatus_0']=mstatus['mstatus_0'].astype('category')
mstatus['mstatus_1']=mstatus['mstatus_1'].astype('category')
mstatus['mstatus_2']=mstatus['mstatus_2'].astype('category')

# add back to original data matrix
data=pd.concat([mstatus,data],axis=1)
data

Seperate y and x matrix, drop vaiables not needed¶

map continuous y to category (change regression to classification)¶

data['pheno_cat']=data['pheno'].map( lambda s : 1 if s < 20 else 0 )
y=data['pheno_cat']
gene=data['gene']
x=data.drop(['gene','pheno','pheno_cat','mstatus'],axis=1)
x_num=data.drop(['gene','pheno','pheno_cat','mstatus','mstatus_0','mstatus_1','mstatus_2'],axis=1)
matrix=data.drop(['gene','pheno','mstatus'],axis=1)

Data explore¶

x.info() # check whether have NAs

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 17 entries, 0 to 16
Data columns (total 90 columns):
mstatus_0    17 non-null uint8
mstatus_1    17 non-null uint8
mstatus_2    17 non-null uint8
gene1        17 non-null float64
gene2        17 non-null float64
gene3        17 non-null float64
gene4        17 non-null float64
gene5        17 non-null float64
gene6        17 non-null float64
gene7        17 non-null float64
gene8        17 non-null float64
gene9        17 non-null float64
gene10       17 non-null float64
gene11       17 non-null float64
gene12       17 non-null float64
gene13       17 non-null float64
gene14       17 non-null float64
gene15       17 non-null float64
gene16       17 non-null float64
gene17       17 non-null float64
gene18       17 non-null float64
gene19       17 non-null float64
gene20       17 non-null float64
gene21       17 non-null float64
gene22       17 non-null float64
gene23       17 non-null float64
gene24       17 non-null float64
gene25       17 non-null float64
gene26       17 non-null float64
gene27       17 non-null float64
gene28       17 non-null float64
gene29       17 non-null float64
gene30       17 non-null float64
gene31       17 non-null float64
gene32       17 non-null float64
gene33       17 non-null float64
gene34       17 non-null float64
gene35       17 non-null float64
gene36       17 non-null float64
gene37       17 non-null float64
gene38       17 non-null float64
gene39       17 non-null float64
gene40       17 non-null float64
gene41       17 non-null float64
gene42       17 non-null float64
gene43       17 non-null float64
gene44       17 non-null float64
gene45       17 non-null float64
gene46       17 non-null float64
gene47       17 non-null float64
gene48       17 non-null float64
gene49       17 non-null float64
gene50       17 non-null float64
gene51       17 non-null float64
gene52       17 non-null float64
gene53       17 non-null float64
gene54       17 non-null float64
gene55       17 non-null float64
gene56       17 non-null float64
gene57       17 non-null float64
gene58       17 non-null float64
gene59       17 non-null float64
gene60       17 non-null float64
gene61       17 non-null float64
gene62       17 non-null float64
gene63       17 non-null float64
gene64       17 non-null float64
gene65       17 non-null float64
gene66       17 non-null float64
gene67       17 non-null float64
gene68       17 non-null float64
gene69       17 non-null float64
gene70       17 non-null float64
gene71       17 non-null float64
gene72       17 non-null float64
gene73       17 non-null float64
gene74       17 non-null float64
gene75       17 non-null float64
gene76       17 non-null float64
gene77       17 non-null float64
gene78       17 non-null float64
gene79       17 non-null float64
gene80       17 non-null float64
gene81       17 non-null float64
gene82       17 non-null float64
gene83       17 non-null float64
gene84       17 non-null float64
gene85       17 non-null float64
gene86       17 non-null float64
gene87       17 non-null float64
dtypes: float64(87), uint8(3)
memory usage: 11.7 KB

x['mstatus_0']=x['mstatus_0'].astype('category')
x['mstatus_1']=x['mstatus_1'].astype('category')
x['mstatus_2']=x['mstatus_2'].astype('category')

x.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 17 entries, 0 to 16
Data columns (total 90 columns):
mstatus_0    17 non-null category
mstatus_1    17 non-null category
mstatus_2    17 non-null category
gene1        17 non-null float64
gene2        17 non-null float64
gene3        17 non-null float64
gene4        17 non-null float64
gene5        17 non-null float64
gene6        17 non-null float64
gene7        17 non-null float64
gene8        17 non-null float64
gene9        17 non-null float64
gene10       17 non-null float64
gene11       17 non-null float64
gene12       17 non-null float64
gene13       17 non-null float64
gene14       17 non-null float64
gene15       17 non-null float64
gene16       17 non-null float64
gene17       17 non-null float64
gene18       17 non-null float64
gene19       17 non-null float64
gene20       17 non-null float64
gene21       17 non-null float64
gene22       17 non-null float64
gene23       17 non-null float64
gene24       17 non-null float64
gene25       17 non-null float64
gene26       17 non-null float64
gene27       17 non-null float64
gene28       17 non-null float64
gene29       17 non-null float64
gene30       17 non-null float64
gene31       17 non-null float64
gene32       17 non-null float64
gene33       17 non-null float64
gene34       17 non-null float64
gene35       17 non-null float64
gene36       17 non-null float64
gene37       17 non-null float64
gene38       17 non-null float64
gene39       17 non-null float64
gene40       17 non-null float64
gene41       17 non-null float64
gene42       17 non-null float64
gene43       17 non-null float64
gene44       17 non-null float64
gene45       17 non-null float64
gene46       17 non-null float64
gene47       17 non-null float64
gene48       17 non-null float64
gene49       17 non-null float64
gene50       17 non-null float64
gene51       17 non-null float64
gene52       17 non-null float64
gene53       17 non-null float64
gene54       17 non-null float64
gene55       17 non-null float64
gene56       17 non-null float64
gene57       17 non-null float64
gene58       17 non-null float64
gene59       17 non-null float64
gene60       17 non-null float64
gene61       17 non-null float64
gene62       17 non-null float64
gene63       17 non-null float64
gene64       17 non-null float64
gene65       17 non-null float64
gene66       17 non-null float64
gene67       17 non-null float64
gene68       17 non-null float64
gene69       17 non-null float64
gene70       17 non-null float64
gene71       17 non-null float64
gene72       17 non-null float64
gene73       17 non-null float64
gene74       17 non-null float64
gene75       17 non-null float64
gene76       17 non-null float64
gene77       17 non-null float64
gene78       17 non-null float64
gene79       17 non-null float64
gene80       17 non-null float64
gene81       17 non-null float64
gene82       17 non-null float64
gene83       17 non-null float64
gene84       17 non-null float64
gene85       17 non-null float64
gene86       17 non-null float64
gene87       17 non-null float64
dtypes: category(3), float64(87)
memory usage: 12.0 KB

data.describe() # check distribution

visualization of variables and features¶

distribution plot¶

import seaborn as sns

sns.distplot(y)

<matplotlib.axes._subplots.AxesSubplot at 0x113d95150>

Heatmap and correlation coefficient¶

corr = matrix.corr()

print (corr['pheno_cat'].sort_values(ascending=False)[:11], '\n')
print (corr['pheno_cat'].sort_values(ascending=False)[-10:])

(pheno_cat    1.000000
gene22       0.703147
gene79       0.635632
gene81       0.604509
gene42       0.589438
gene3        0.585379
gene62       0.576375
gene80       0.572827
gene47       0.551429
gene66       0.525566
gene15       0.522567
Name: pheno_cat, dtype: float64, '\n')
gene48      -0.163747
gene12      -0.219666
gene28      -0.230907
gene33      -0.240243
gene69      -0.240599
gene4       -0.242297
gene18      -0.256976
gene20      -0.259309
gene84      -0.268074
mstatus_0   -0.409673
Name: pheno_cat, dtype: float64

def plot_correlation_map( df ):
    corr = matrix.corr()
    _ , ax = plt.subplots( figsize =( 12 , 10 ) )
    cmap = sns.diverging_palette( 220 , 10 , as_cmap = True )
    _ = sns.heatmap(
        corr, 
        cmap = cmap,
        square=True, 
        cbar_kws={ 'shrink' : .9 }, 
        ax=ax, 
        #annot = True, 
        #annot_kws = { 'fontsize' : 12 }
    )

plot_correlation_map( matrix )

Standardize numerical variables¶

# no need for standardization for random forest classification

x_num=data.drop(['gene','pheno','pheno_cat','mstatus','mstatus_0','mstatus_1','mstatus_2'],axis=1)

from sklearn.preprocessing import StandardScaler
scalerX = StandardScaler().fit(x_num)
x_std = scalerX.transform(x_num)
#print x_std

col_names = list(x_num)
print col_names
x_features = pd.DataFrame(x_std, columns = col_names)
x_features.head()

['gene1', 'gene2', 'gene3', 'gene4', 'gene5', 'gene6', 'gene7', 'gene8', 'gene9', 'gene10', 'gene11', 'gene12', 'gene13', 'gene14', 'gene15', 'gene16', 'gene17', 'gene18', 'gene19', 'gene20', 'gene21', 'gene22', 'gene23', 'gene24', 'gene25', 'gene26', 'gene27', 'gene28', 'gene29', 'gene30', 'gene31', 'gene32', 'gene33', 'gene34', 'gene35', 'gene36', 'gene37', 'gene38', 'gene39', 'gene40', 'gene41', 'gene42', 'gene43', 'gene44', 'gene45', 'gene46', 'gene47', 'gene48', 'gene49', 'gene50', 'gene51', 'gene52', 'gene53', 'gene54', 'gene55', 'gene56', 'gene57', 'gene58', 'gene59', 'gene60', 'gene61', 'gene62', 'gene63', 'gene64', 'gene65', 'gene66', 'gene67', 'gene68', 'gene69', 'gene70', 'gene71', 'gene72', 'gene73', 'gene74', 'gene75', 'gene76', 'gene77', 'gene78', 'gene79', 'gene80', 'gene81', 'gene82', 'gene83', 'gene84', 'gene85', 'gene86', 'gene87']

x_features=pd.concat([mstatus,x_features],axis=1)
x_features

y_std=y

Seperate data into train and test¶

from sklearn.cross_validation import *

def train_and_evaluate(clf, X_train, y_train):
    cv = KFold(X_train.shape[0], 5, shuffle=True, random_state=33)
    scores = cross_val_score(clf, X_train, y_train, cv=cv)
    print ('Average coefficient of determination using 5-fold cross validation:', np.mean(scores))

/Users/shanshanyang/anaconda/envs/gl-env/lib/python2.7/site-packages/sklearn/cross_validation.py:44: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.
  "This module will be removed in 0.20.", DeprecationWarning)

Feature selection¶

modeling before feature selection¶

from sklearn.ensemble import RandomForestClassifier
rfc = RandomForestClassifier(max_depth=3, min_samples_leaf=5,n_estimators=100)
train_and_evaluate(rfc,x,y)

('Average coefficient of determination using 5-fold cross validation:', 0.66666666666666663)

train_and_evaluate(rfc,x_features,y_std)

('Average coefficient of determination using 5-fold cross validation:', 0.66666666666666663)

Feature selection begin¶

Feature selection by random forest importance¶

from sklearn.ensemble import RandomForestClassifier

feat_labels = x.columns[0:]

forest = RandomForestClassifier(n_estimators=100,
                                random_state=0,
                                n_jobs=-1)

forest.fit(x, y)
importances = forest.feature_importances_

indices = np.argsort(importances)[::-1]

for f in range(x_features.shape[1]):
    print("%2d) %-*s %f" % (f + 1, 30, 
                            feat_labels[indices[f]], 
                            importances[indices[f]]))

plt.title('Feature Importances')
plt.bar(range(x.shape[1]), 
        importances[indices],
        color='lightblue', 
        align='center')

plt.xticks(range(x.shape[1]), 
           feat_labels[indices], rotation=90)
plt.xlim([-1, x.shape[1]])
plt.tight_layout()
#plt.savefig('./random_forest.png', dpi=300)
plt.show()

 1) gene80                         0.064684
 2) gene11                         0.044370
 3) gene7                          0.041469
 4) gene42                         0.040682
 5) gene22                         0.037366
 6) gene79                         0.036462
 7) gene69                         0.031680
 8) gene39                         0.028793
 9) gene62                         0.027032
10) gene72                         0.024817
11) gene23                         0.022711
12) gene31                         0.020742
13) gene46                         0.020489
14) gene55                         0.020398
15) gene71                         0.020335
16) gene53                         0.019974
17) gene86                         0.019755
18) gene6                          0.019231
19) gene81                         0.018288
20) gene51                         0.017927
21) gene17                         0.017908
22) gene47                         0.017875
23) gene78                         0.017792
24) gene66                         0.017622
25) gene68                         0.016577
26) gene35                         0.014913
27) gene50                         0.014570
28) gene40                         0.014220
29) gene34                         0.013088
30) gene52                         0.012367
31) gene26                         0.012267
32) gene5                          0.012185
33) gene37                         0.012066
34) gene8                          0.011503
35) gene28                         0.011163
36) gene57                         0.010828
37) gene56                         0.010495
38) gene30                         0.010117
39) gene84                         0.010000
40) gene38                         0.009746
41) gene64                         0.009505
42) gene59                         0.008905
43) gene60                         0.008583
44) gene73                         0.008077
45) gene49                         0.007792
46) gene36                         0.007740
47) gene63                         0.007210
48) gene12                         0.006222
49) gene3                          0.005916
50) gene67                         0.005733
51) gene58                         0.005641
52) gene13                         0.005641
53) gene32                         0.005559
54) gene27                         0.005380
55) gene48                         0.004813
56) gene18                         0.004808
57) gene1                          0.004530
58) gene54                         0.004359
59) gene61                         0.004333
60) gene75                         0.003279
61) gene20                         0.003036
62) gene19                         0.002361
63) gene87                         0.002361
64) gene16                         0.002208
65) gene45                         0.002208
66) gene82                         0.002186
67) gene21                         0.002186
68) gene14                         0.002099
69) gene4                          0.002099
70) gene83                         0.002082
71) gene74                         0.002082
72) gene44                         0.002066
73) gene77                         0.001889
74) gene15                         0.001635
75) gene2                          0.000971
76) gene76                         0.000000
77) gene85                         0.000000
78) mstatus_2                      0.000000
79) mstatus_1                      0.000000
80) gene24                         0.000000
81) gene70                         0.000000
82) gene9                          0.000000
83) gene10                         0.000000
84) gene25                         0.000000
85) gene65                         0.000000
86) gene43                         0.000000
87) gene41                         0.000000
88) gene33                         0.000000
89) gene29                         0.000000
90) mstatus_0                      0.000000

forest = forest.fit(x, y)

from sklearn.feature_selection import SelectFromModel
selector = SelectFromModel(forest, prefit=True)
x_subsets_0 = selector.transform(x)
idxs_selected=selector.get_support()
labels = x.columns[idxs_selected]
x_subsets_0=pd.DataFrame(x_subsets_0, columns=labels)
x_subsets_0
print labels

Index([u'gene5', u'gene6', u'gene7', u'gene8', u'gene11', u'gene17', u'gene22',
       u'gene23', u'gene26', u'gene28', u'gene31', u'gene34', u'gene35',
       u'gene37', u'gene39', u'gene40', u'gene42', u'gene46', u'gene47',
       u'gene50', u'gene51', u'gene52', u'gene53', u'gene55', u'gene62',
       u'gene66', u'gene68', u'gene69', u'gene71', u'gene72', u'gene78',
       u'gene79', u'gene80', u'gene81', u'gene86'],
      dtype='object')

x_subsets_0

Get feature subsets by chi2_KBest¶

from sklearn.feature_selection import SelectKBest
from sklearn.feature_selection import chi2

#x_select=SelectKBest(chi2, k=20).fit_transform(x,y)
def KBest_selector(x,y):
    #columns = data.columns
    selector = SelectKBest(chi2, k=20)
    selector.fit_transform(x,y)
    idxs_selected=selector.get_support()
    labels = x.columns[idxs_selected]
    print labels
    return pd.DataFrame(selector.fit_transform(x,y), columns=labels)
x_subsets_3=KBest_selector(x_subsets_0,y)

Index([u'gene8', u'gene22', u'gene26', u'gene34', u'gene35', u'gene37',
       u'gene39', u'gene42', u'gene46', u'gene47', u'gene50', u'gene51',
       u'gene53', u'gene55', u'gene66', u'gene68', u'gene78', u'gene79',
       u'gene80', u'gene86'],
      dtype='object')

x_subsets_3.shape

(17, 20)

Assign new data matrix¶

x_subsets_3

subset_matrix=pd.concat([gene,y,mstatus,x_subsets_3],axis=1)
subset_matrix_1=pd.concat([mstatus,x_subsets_3],axis=1)
subset_matrix

Clustering by features¶

from scipy.spatial.distance import pdist, squareform
import numpy as np
labels=list(subset_matrix['pheno_cat'])
print labels

row_dist = pdist(subset_matrix_1, metric='euclidean')
row_dist

from scipy.cluster.hierarchy import linkage, dendrogram
row_clusters = linkage(row_dist, method='complete', metric='euclidean')

# make dendrogram black (part 1/2)
# from scipy.cluster.hierarchy import set_link_color_palette
# set_link_color_palette(['black'])

row_dendr = dendrogram(row_clusters, 
                       labels=labels,
                       orientation='left'
                       # make dendrogram black (part 2/2)
                       # color_threshold=np.inf
                       )
plt.tight_layout()
plt.ylabel('Euclidean distance')
#plt.savefig('./figures/dendrogram.png', dpi=300, 
#            bbox_inches='tight')
plt.show()

[0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 1]

x_subsets_new_allm=pd.concat([mstatus,x_subsets_3],axis=1)
x_subsets_new_allm

Modeling after feature selections¶

Random Forest (no need for standrization)¶

from sklearn.ensemble import RandomForestClassifier
rfc = RandomForestClassifier(max_depth=3, min_samples_leaf=5,n_estimators=100)

train_and_evaluate(rfc,x_subsets_new_allm,y)

('Average coefficient of determination using 5-fold cross validation:', 0.66666666666666663)

SVM¶

standardize features¶

from sklearn.preprocessing import StandardScaler
scalerX = StandardScaler().fit(x_subsets_3)
x_std = scalerX.transform(x_subsets_3)

col_names = list(x_subsets_3)
print col_names
x_subsets_std = pd.DataFrame(x_std, columns = col_names)
x_subsets_std.head()

['gene8', 'gene22', 'gene26', 'gene34', 'gene35', 'gene37', 'gene39', 'gene42', 'gene46', 'gene47', 'gene50', 'gene51', 'gene53', 'gene55', 'gene66', 'gene68', 'gene78', 'gene79', 'gene80', 'gene86']

x_subsets_std_new_allm=pd.concat([mstatus,x_subsets_std],axis=1)

from sklearn import svm
svc_linear = svm.SVC(kernel='linear')
train_and_evaluate(svc_linear,x_subsets_std_new_allm,y)

('Average coefficient of determination using 5-fold cross validation:', 0.71666666666666656)

Apply model on test data sets¶

svc_linear.score(x_test,y_test) # just example

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
<ipython-input-44-1490464f8ace> in <module>()
----> 1 svc_linear.score(x_test,y_test)

NameError: name 'x_test' is not defined

	gene	pheno	mstatus	gene1	gene2	gene3	gene4	gene5	gene6	gene7	...	gene78	gene79	gene80	gene81	gene82	gene83	gene84	gene85	gene86	gene87
0	cellline1	102.800	2	35.563389	12618.769920	0.380812	83.539161	24616.88273	507.121965	33.113287	...	1664.275029	9884.085946	9967.348993	0.180494	1925.264711	21447.89539	440.137272	8334.957054	10.725748	29.126002
1	cellline2	1.007	2	2.412029	15493.175530	324.463153	81.020538	11889.10427	307.848156	43.459984	...	2102.489421	20784.919950	11924.561280	17.737673	1624.761674	17661.26251	8.204143	4742.916507	123.666991	28.482386
2	cellline3	248.300	1	11.895738	4164.217113	2.609915	180.803666	22294.39146	359.055020	14.090481	...	1430.476500	12276.371810	7596.776712	1.253727	189.370291	11493.70647	578.183845	5957.650724	17.089377	44.979818
3	cellline4	8.200	2	10.841646	11167.453320	619.654523	1.607177	39831.41843	409.473237	0.051204	...	1645.797712	15493.633420	14700.996270	18.169115	514.287852	14091.65444	2.374032	4547.726810	120.232935	8.231084
4	cellline5	9.140	1	35.091377	6418.275182	1051.428133	113.379095	11214.43702	516.293687	175.544484	...	2521.399818	15978.260150	12423.272770	17.881719	1228.890831	17383.92604	74.218717	9324.497865	114.755041	58.974125

	mstatus_0	mstatus_1	mstatus_2	gene	pheno	mstatus	gene1	gene2	gene3	gene4	...	gene78	gene79	gene80	gene81	gene82	gene83	gene84	gene85	gene86	gene87
0	0	0	1	cellline1	102.800	2	35.563389	12618.769920	0.380812	83.539161	...	1664.275029	9884.085946	9967.348993	0.180494	1925.264711	21447.89539	440.137272	8334.957054	10.725748	29.126002
1	0	0	1	cellline2	1.007	2	2.412029	15493.175530	324.463153	81.020538	...	2102.489421	20784.919950	11924.561280	17.737673	1624.761674	17661.26251	8.204143	4742.916507	123.666991	28.482386
2	0	1	0	cellline3	248.300	1	11.895738	4164.217113	2.609915	180.803666	...	1430.476500	12276.371810	7596.776712	1.253727	189.370291	11493.70647	578.183845	5957.650724	17.089377	44.979818
3	0	0	1	cellline4	8.200	2	10.841646	11167.453320	619.654523	1.607177	...	1645.797712	15493.633420	14700.996270	18.169115	514.287852	14091.65444	2.374032	4547.726810	120.232935	8.231084
4	0	1	0	cellline5	9.140	1	35.091377	6418.275182	1051.428133	113.379095	...	2521.399818	15978.260150	12423.272770	17.881719	1228.890831	17383.92604	74.218717	9324.497865	114.755041	58.974125
5	0	0	1	cellline6	110.400	2	17.183818	12482.024900	269.561811	51.533824	...	706.746609	7576.531063	4248.833977	14.982130	681.371465	8793.60973	1.656686	3500.790305	15.520336	55.060588
6	1	0	0	cellline7	41.880	0	0.371414	20549.792360	2.487057	307.895894	...	1290.303312	11198.139500	9234.434085	0.058224	730.706882	12951.75096	456.451955	5240.821567	3.955019	72.644239
7	0	1	0	cellline8	63.390	1	45.146636	7189.947832	411.563472	178.920283	...	1617.006495	12001.325160	11649.730500	0.517588	663.610145	21936.46752	74.244456	6125.073377	183.751932	18.223628
8	0	0	1	cellline9	75.680	2	11.990382	9149.985952	164.285280	94.617134	...	1428.388276	10226.198160	9327.724256	0.886400	591.754297	13540.44950	74.325502	4974.957898	0.310012	96.512793
9	0	0	1	cellline10	44.050	2	11.626690	11830.781570	0.358404	76.789411	...	628.281564	6629.017964	3095.952456	4.479230	686.023868	19208.27344	431.874493	3349.225674	11.499536	75.489329
10	1	0	0	cellline11	281.190	0	2.132654	10976.700230	0.862692	109.289518	...	1207.735707	7361.351131	3089.205933	4.346857	403.807281	7460.05376	1120.277218	2823.917819	11.162283	34.136050
11	0	0	1	cellline12	9.570	2	12.858604	15689.729750	796.332679	0.132118	...	950.680843	14052.822460	11551.979880	20.425984	1111.175871	13747.30643	0.203818	5268.234115	112.458372	35.010884
12	0	0	1	cellline13	49.090	2	19.568435	10137.978600	0.274555	91.518662	...	597.463348	8393.873894	5697.246205	1.622008	632.169031	23452.01991	574.132513	1468.491177	14.927268	63.143617
13	1	0	0	cellline14	290.400	0	13.970087	16582.077530	1.881488	192.808114	...	2263.587397	18489.697680	8242.025026	0.186814	1129.169580	13685.21405	372.551732	4105.246853	9.003845	27.135906
14	0	0	1	cellline15	10.960	2	24.595825	7488.448799	6.323575	242.627061	...	1102.945580	10715.592810	7245.659314	0.078100	319.394816	17372.60156	916.370833	3982.293889	40.324223	80.732546
15	1	0	0	cellline16	105.920	0	20.859663	21835.587420	2.378412	90.958827	...	869.249267	12152.991810	10004.144460	0.000000	668.819973	25722.49928	537.708645	4085.655646	5.817749	59.507186
16	0	0	1	cellline17	8.870	2	29.573701	9220.727051	1.210296	120.047963	...	1981.951718	20464.962880	10268.070270	0.225657	1058.295671	12194.77736	447.381173	5014.393265	9.473663	98.453777

	mstatus_0	mstatus_1	mstatus_2	pheno	mstatus	gene1	gene2	gene3	gene4	gene5	...	gene79	gene80	gene81	gene82	gene83	gene84	gene85	gene86	gene87	pheno_cat
count	17.000000	17.000000	17.000000	17.000000	17.000000	17.000000	17.000000	17.000000	17.000000	17.000000	...	17.000000	17.000000	17.000000	17.000000	17.000000	17.000000	17.000000	17.000000	17.000000	17.000000
mean	0.235294	0.176471	0.588235	85.932176	1.352941	17.981299	11940.921945	215.062133	118.675791	18508.949853	...	12569.398576	8839.291905	6.060689	832.874955	16008.439315	359.429237	4873.344150	47.333784	52.108468	0.352941
std	0.437237	0.392953	0.507300	96.755571	0.861770	12.598904	4836.929875	327.081614	80.245059	8942.765880	...	4406.193937	3339.915409	8.008130	458.638295	5128.941717	335.013301	1887.238831	58.210639	26.913396	0.492592
min	0.000000	0.000000	0.000000	1.007000	0.000000	0.371414	4164.217113	0.274555	0.132118	5032.240744	...	6629.017964	3089.205933	0.000000	189.370291	7460.053760	0.203818	1468.491177	0.310012	8.231084	0.000000
25%	0.000000	0.000000	0.000000	9.570000	1.000000	11.626690	9149.985952	1.210296	81.020538	11214.437020	...	9884.085946	7245.659314	0.186814	591.754297	12951.750960	74.218717	3982.293889	9.473663	29.126002	0.000000
50%	0.000000	0.000000	1.000000	49.090000	2.000000	13.970087	11167.453320	2.609915	94.617134	18834.957130	...	12001.325160	9327.724256	1.253727	681.371465	14091.654440	431.874493	4742.916507	14.927268	55.060588	0.000000
75%	0.000000	0.000000	1.000000	105.920000	2.000000	24.595825	15493.175530	324.463153	178.920283	23444.097860	...	15493.633420	11551.979880	14.982130	1111.175871	19208.273440	537.708645	5268.234115	112.458372	72.644239	1.000000
max	1.000000	1.000000	1.000000	290.400000	2.000000	45.146636	21835.587420	1051.428133	307.895894	39831.418430	...	20784.919950	14700.996270	20.425984	1925.264711	25722.499280	1120.277218	9324.497865	183.751932	98.453777	1.000000

	gene1	gene2	gene3	gene4	gene5	gene6	gene7	gene8	gene9	gene10	...	gene78	gene79	gene80	gene81	gene82	gene83	gene84	gene85	gene86	gene87
0	1.438474	0.144453	-0.676554	-0.451343	0.704023	1.235369	-0.250141	-0.262056	-0.255102	2.331508	...	0.449070	-0.628197	0.348145	-0.756877	2.455115	1.093181	0.248324	1.890672	-0.648244	-0.880223
1	-1.273796	0.757005	0.344770	-0.483695	-0.763028	0.047726	-0.147672	-0.249421	-0.255579	-0.383586	...	1.229998	1.921923	0.952187	1.503017	1.779743	0.332172	-1.080657	-0.071237	1.351685	-0.904873
2	-0.497889	-1.657259	-0.669529	0.798055	0.436324	0.352911	-0.438533	-0.273704	-0.276348	-1.116267	...	0.032425	-0.068550	-0.383469	-0.618734	-1.446258	-0.907337	0.673069	0.592229	-0.535559	-0.273026
3	-0.584129	-0.164830	1.275047	-1.503788	2.457707	0.653397	-0.577571	-0.133307	-0.275905	-0.253917	...	0.416142	0.684090	1.809060	1.558551	-0.716015	-0.385221	-1.098595	-0.177846	1.290876	-1.680493
4	1.399857	-1.176906	2.635753	-0.068038	-0.840793	1.290031	1.160428	3.997331	3.994326	-1.045703	...	1.976525	0.797462	1.106101	1.521559	0.890034	0.276435	-0.877542	2.431141	1.193875	0.262953

	mstatus_0	mstatus_1	mstatus_2	gene1	gene2	gene3	gene4	gene5	gene6	gene7	...	gene78	gene79	gene80	gene81	gene82	gene83	gene84	gene85	gene86	gene87
0	0	0	1	1.438474	0.144453	-0.676554	-0.451343	0.704023	1.235369	-0.250141	...	0.449070	-0.628197	0.348145	-0.756877	2.455115	1.093181	0.248324	1.890672	-0.648244	-0.880223
1	0	0	1	-1.273796	0.757005	0.344770	-0.483695	-0.763028	0.047726	-0.147672	...	1.229998	1.921923	0.952187	1.503017	1.779743	0.332172	-1.080657	-0.071237	1.351685	-0.904873
2	0	1	0	-0.497889	-1.657259	-0.669529	0.798055	0.436324	0.352911	-0.438533	...	0.032425	-0.068550	-0.383469	-0.618734	-1.446258	-0.907337	0.673069	0.592229	-0.535559	-0.273026
3	0	0	1	-0.584129	-0.164830	1.275047	-1.503788	2.457707	0.653397	-0.577571	...	0.416142	0.684090	1.809060	1.558551	-0.716015	-0.385221	-1.098595	-0.177846	1.290876	-1.680493
4	0	1	0	1.399857	-1.176906	2.635753	-0.068038	-0.840793	1.290031	1.160428	...	1.976525	0.797462	1.106101	1.521559	0.890034	0.276435	-0.877542	2.431141	1.193875	0.262953
5	0	0	1	-0.065246	0.115312	0.171752	-0.862463	-1.553376	-0.734681	-0.562538	...	-1.257310	-1.168022	-1.416723	1.148334	-0.340500	-1.449983	-1.100803	-0.749665	-0.563343	0.113065
6	1	0	0	-1.440749	1.834598	-0.669917	2.430600	-1.317385	-0.693931	0.138479	...	-0.217373	-0.320790	0.121950	-0.772615	-0.229620	-0.614310	0.298522	0.200710	-0.768138	0.786515
7	0	1	0	2.222526	-1.012459	0.619261	0.773862	0.827537	2.538710	3.133539	...	0.364835	-0.132894	0.867367	-0.713487	-0.380418	1.191371	-0.877463	0.683672	2.415651	-1.297781
8	0	0	1	-0.490145	-0.594764	-0.160020	-0.309042	1.318473	0.752538	1.656796	...	0.028704	-0.548164	0.150742	-0.666015	-0.541912	-0.495998	-0.877214	0.055500	-0.832683	1.700675
9	0	0	1	-0.519901	-0.023472	-0.676625	-0.538045	-1.075514	-1.023535	-0.418426	...	-1.397140	-1.389681	-1.772530	-0.203560	-0.330044	0.643079	0.222901	-0.832447	-0.634542	0.895481
10	1	0	0	-1.296653	-0.205481	-0.675036	-0.120570	-0.159645	-0.962268	-0.490681	...	-0.364514	-1.218360	-1.774612	-0.220598	-0.964317	-1.717991	2.340994	-1.119360	-0.640514	-0.688339
11	0	0	1	-0.419112	0.798892	1.831836	-1.522736	-0.257609	0.091208	-0.578078	...	-0.822603	0.347029	0.837199	1.849047	0.625473	-0.454426	-1.105273	0.215682	1.153207	-0.654833
12	0	0	1	0.129851	-0.384217	-0.676889	-0.348843	0.037577	-1.305151	-0.502113	...	-1.452061	-0.976814	-0.969709	-0.571330	-0.451081	1.495955	0.660604	-1.859670	-0.573845	0.422643
13	1	0	0	-0.328176	0.989056	-0.671825	0.952256	0.568843	-0.760771	-0.431594	...	1.517086	1.384983	-0.184331	-0.756063	0.665914	-0.466904	0.040376	-0.419521	-0.678735	-0.956443
14	0	0	1	0.541166	-0.948847	-0.657826	1.592198	0.215315	-0.377524	-0.574022	...	-0.551257	-0.433676	-0.491832	-0.770056	-1.154032	0.274159	1.713610	-0.486676	-0.124123	1.096295
15	1	0	0	0.235493	2.108608	-0.670259	-0.356034	0.254775	-0.873898	-0.539792	...	-0.967720	-0.097413	0.359501	-0.780109	-0.368709	1.952259	0.548534	-0.430222	-0.735153	0.283369
16	0	0	1	0.948430	-0.579689	-0.673940	0.017626	-0.853225	-0.230129	-0.578078	...	1.015192	1.847073	0.440955	-0.751063	0.506627	-0.766441	0.270612	0.077039	-0.670416	1.775014

	mstatus_0	mstatus_1	mstatus_2
0	0	0	1
1	0	0	1
2	0	1	0
3	0	0	1
4	0	1	0
5	0	0	1
6	1	0	0
7	0	1	0
8	0	0	1
9	0	0	1
10	1	0	0
11	0	0	1
12	0	0	1
13	1	0	0
14	0	0	1
15	1	0	0
16	0	0	1

	gene5	gene6	gene7	gene8	gene11	gene17	gene22	gene23	gene26	gene28	...	gene66	gene68	gene69	gene71	gene72	gene78	gene79	gene80	gene81	gene86
0	24616.882730	507.121965	33.113287	1.872366	0.000000	355.992343	381.771416	112.540158	7.155964	0.000000	...	1829.559917	48.724483	54.666330	18.990077	0.932670	1664.275029	9884.085946	9967.348993	0.180494	10.725748
1	11889.104270	307.848156	43.459984	3.903626	3.728302	280.125569	461.684201	51.501824	7.187581	7.414297	...	1721.916413	171.617876	0.000000	1.049774	7.574311	2102.489421	20784.919950	11924.561280	17.737673	123.666991
2	22294.391460	359.055020	14.090481	0.000000	0.754708	99.338788	276.901237	168.440576	3.368073	1.931817	...	1156.755256	13.214744	31.580594	16.840515	0.160708	1430.476500	12276.371810	7596.776712	1.253727	17.089377
3	39831.418430	409.473237	0.051204	22.569492	3.625660	174.663149	385.075970	76.421466	114.926339	1.148442	...	1633.818172	125.919430	2.952763	0.261735	33.236943	1645.797712	15493.633420	14700.996270	18.169115	120.232935
4	11214.437020	516.293687	175.544484	686.590180	26.752111	286.468321	506.957655	93.836650	99.341986	3.144239	...	1549.144961	143.048764	6.989222	251.183858	6.675416	2521.399818	15978.260150	12423.272770	17.881719	114.755041
5	5032.240744	176.568608	1.569169	10.710532	0.000000	424.437755	145.365462	47.901209	50.646195	0.262773	...	1982.789612	147.689713	39.339566	5.641386	4.347309	706.746609	7576.531063	4248.833977	14.982130	15.520336
6	7079.635161	183.406174	72.353869	0.000000	0.252355	169.262727	225.968047	97.691572	2.570058	5.694864	...	376.790338	7.842055	29.375955	49.965914	0.312417	1290.303312	11198.139500	9234.434085	0.058224	3.955019
7	25688.460820	725.808594	374.778085	7.968790	13.634660	325.341969	248.441013	40.647865	24.252006	18.636545	...	986.034243	272.998199	6.875733	190.474292	0.180915	1617.006495	12001.325160	11649.730500	0.517588	183.751932
8	29947.698990	426.108074	225.664922	4.737565	3.501571	142.726975	214.042162	43.229987	20.633431	30.134704	...	1557.842682	56.301952	9.421854	390.440028	4.307026	1428.388276	10226.198160	9327.724256	0.886400	0.310012
9	9178.049873	128.102024	16.120752	0.000000	1.611854	239.985693	74.999562	20.108970	0.568929	0.955794	...	1901.177614	78.588255	121.044681	0.156302	0.000000	628.281564	6629.017964	3095.952456	4.479230	11.499536
10	17123.908130	138.382013	8.824945	0.000000	1.188597	120.951173	152.624437	49.897361	3.352154	3.773412	...	1056.202929	25.261415	82.665247	19.483341	1.115975	1207.735707	7361.351131	3089.205933	4.346857	11.162283
11	16273.998640	315.143989	0.000000	9.634266	19.223147	348.098780	404.388349	86.235939	20.653271	0.000000	...	2908.369075	389.631949	0.351922	15.159488	5.370214	950.680843	14052.822460	11551.979880	20.425984	112.458372
12	18834.957130	80.849918	7.670536	0.000000	0.046837	264.073617	203.616439	67.634926	3.645116	0.138355	...	1347.081268	27.344872	201.398403	24.037126	0.000000	597.463348	8393.873894	5697.246205	1.622008	14.927268
13	23444.097860	172.191121	14.791154	0.000000	0.066646	238.122411	299.734488	71.132905	2.551197	0.196870	...	702.386186	36.043874	11.323641	30.883144	0.000000	2263.587397	18489.697680	8242.025026	0.186814	9.003845
14	20376.972130	236.495778	0.409571	0.000000	0.000000	83.913846	183.937425	227.811017	0.000000	0.000000	...	1163.908426	0.149122	127.031139	4.072595	0.000000	1102.945580	10715.592810	7245.659314	0.078100	40.324223
15	20719.316530	153.209612	3.865932	0.000000	3.499002	134.238635	117.436898	56.215857	1.494275	1.418896	...	566.466533	33.319854	54.029393	4.878767	0.000000	869.249267	12152.991810	10004.144460	0.000000	5.817749
16	11106.577590	261.227059	0.000000	0.000000	8.386553	425.999096	502.194685	115.709317	2.988747	0.000000	...	2869.240281	23.278064	49.061018	10.967131	0.000000	1981.951718	20464.962880	10268.070270	0.225657	9.473663

	gene8	gene22	gene26	gene34	gene35	gene37	gene39	gene42	gene46	gene47	gene50	gene51	gene53	gene55	gene66	gene68	gene78	gene79	gene80	gene86
0	-0.262056	0.758455	-0.424387	-0.196038	-0.443661	-0.503696	0.839409	-0.687246	-0.819300	-0.769308	-0.248128	-0.270435	0.083470	-0.651182	0.499368	-0.442380	0.449070	-0.628197	0.348145	-0.648244
1	-0.249421	1.362790	-0.423451	0.186680	-0.180933	-0.342529	-0.346356	1.719083	0.870488	0.865921	-0.244713	3.990010	2.035050	-0.242747	0.341625	0.753770	1.229998	1.921923	0.952187	1.351685
2	-0.273704	-0.034617	-0.536533	-0.336082	-0.413692	-0.504013	-1.175351	-0.713695	-0.818744	-0.724661	-0.250690	-0.249343	-1.229242	-0.650354	-0.486577	-0.788005	0.032425	-0.068550	-0.383469	-0.535559
3	-0.133307	0.783446	2.766298	-0.688852	-0.421641	2.875857	1.987489	1.441008	1.298080	1.291325	-0.250690	-0.164358	-0.425525	2.429292	0.212523	0.308976	0.416142	0.684090	1.809060	1.290876
4	3.997331	1.705167	2.304902	0.506001	2.620877	0.857807	-0.203392	2.238724	1.759354	2.895414	3.999996	-0.217351	1.589919	0.497005	0.088441	0.475700	1.976525	0.797462	1.106101	1.193875

	mstatus_0	mstatus_1	mstatus_2
0	0	0	1
1	0	0	1
2	0	1	0
3	0	0	1
4	0	1	0
5	0	0	1
6	1	0	0
7	0	1	0
8	0	0	1
9	0	0	1
10	1	0	0
11	0	0	1
12	0	0	1
13	1	0	0
14	0	0	1
15	1	0	0
16	0	0	1

	mstatus_0	mstatus_1	mstatus_2
0	0	0	1
1	0	0	1
2	0	1	0
3	0	0	1
4	0	1	0
5	0	0	1
6	1	0	0
7	0	1	0
8	0	0	1
9	0	0	1
10	1	0	0
11	0	0	1
12	0	0	1
13	1	0	0
14	0	0	1
15	1	0	0
16	0	0	1