In [3]:

    

from concise.data import attract, encode


    

In [50]:

    

from plotnine import *


    

    




/home/avsec/bin/anaconda3/lib/python3.5/site-packages/statsmodels/compat/pandas.py:56: FutureWarning: The pandas.core.datetools module is deprecated and will be removed in a future version. Please use the pandas.tseries module instead.
  from pandas.core import datetools

In [18]:

    

import numpy as np
import pandas as pd


    

In [42]:

    

pwm_list = attract.get_pwm_list(attract.get_metadata().PWM_id.unique())
pwm_list_e = encode.get_pwm_list(encode.get_metadata().motif_name.unique())


    

In [43]:

    

pwm_list[0].pwm.shape


    

    
Out[43]:





(6, 4)

In [44]:

    

pwmc = np.concatenate([pwm.pwm for pwm in pwm_list], axis=0)
pwmce = np.concatenate([pwm.pwm for pwm in pwm_list_e], axis=0)


    

In [46]:

    

dfa = pd.DataFrame(pwmc, columns = ["A", "C", "G", "T"])
dfae = pd.DataFrame(pwmce, columns = ["A", "C", "G", "T"])


    

In [48]:

    

dfa = pd.melt(dfa, var_name="base")
dfa["db"] = "attract"
dfae = pd.melt(dfae, var_name="base")
dfae["db"] = "encode"
df = pd.concat([dfa, dfae])


    

In [49]:

    

df


    

    
Out[49]:







  

    

      

      
base
      
value
      
db
    
  
  

    

      
0
      
A
      
0.0096
      
attract
    
    

      
1
      
A
      
0.9711
      
attract
    
    

      
2
      
A
      
0.9711
      
attract
    
    

      
...
      
...
      
...
      
...
    
    

      
107453
      
T
      
0.0500
      
encode
    
    

      
107454
      
T
      
0.0500
      
encode
    
    

      
107455
      
T
      
0.0500
      
encode
    
  

152328 rows × 3 columns

In [67]:

    

df.groupby("db").count()


    

    
Out[67]:







  

    

      

      
base
      
value
    
    

      
db
      

      

    
  
  

    

      
attract
      
44872
      
44872
    
    

      
encode
      
107456
      
107456
    
  

In [62]:

    

ggplot(aes(x = "value", color = "base"), df) + \
    geom_density() + \
    facet_grid('db~.') + \
    ylim((0, 8))


    

    













    
Out[62]:





<ggplot: (-9223363275156352708)>

In [ ]: