In [1]:

    

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


    

In [2]:

    

path_pro = '/Users/luke/singlecell/krse2013-MT/fpkm_groups_all_pro.binary.Sta3Surf.Gareth.csv'
path_sar = '/Users/luke/singlecell/krse2013-MT/fpkm_groups_all_sar.binary.Sta3Surf.csv'


    

In [3]:

    

df_pro = pd.read_csv(path_pro, index_col=0)
df_sar = pd.read_csv(path_sar, index_col=0)


    

In [4]:

    

df_pro.head()


    

    
Out[4]:






  

    

      

      
4415_Sta3_T0_Surf_20
      
4419_Sta3_T4_Surf_71
      
4423_Sta3_T8_Surf_75
      
4427_Sta3_T12_Surf_79
      
4431_Sta3_T16_Surf_83
      
4435_Sta3_T20_Surf_87
      
4439_Sta3_T24_Surf_91
      
4443_Sta3_T28_Surf_95
      
4447_Sta3_T32_Surf_99
      
4451_Sta3_T36_Surf_103
      
4455_Sta3_T40_Surf_107
      
4459_Sta3_T44_Surf_111
      
4463_Sta3_T48_Surf_115
      
Group
    
    

      
GeneID
      

      

      

      

      

      

      

      

      

      

      

      

      

      

    
  
  

    

      
pro11876
      
0
      
1
      
1
      
1
      
1
      
1
      
1
      
1
      
1
      
1
      
1
      
1
      
1
      
RedSea
    
    

      
pro11957
      
0
      
0
      
1
      
1
      
1
      
1
      
0
      
0
      
1
      
1
      
1
      
1
      
1
      
RedSea
    
    

      
pro11985
      
0
      
1
      
0
      
1
      
1
      
1
      
1
      
1
      
0
      
1
      
1
      
1
      
1
      
RedSea
    
    

      
pro11988
      
0
      
1
      
1
      
1
      
1
      
1
      
1
      
1
      
1
      
1
      
1
      
1
      
1
      
RedSea
    
    

      
pro11989
      
0
      
1
      
1
      
1
      
1
      
1
      
1
      
1
      
1
      
1
      
1
      
1
      
1
      
RedSea
    
  

In [5]:

    

df_pro.Group.value_counts()


    

    
Out[5]:





Other     8691
RedSea     637
Name: Group, dtype: int64

In [6]:

    

df_pro.columns


    

    
Out[6]:





Index(['4415_Sta3_T0_Surf_20', '4419_Sta3_T4_Surf_71', '4423_Sta3_T8_Surf_75',
       '4427_Sta3_T12_Surf_79', '4431_Sta3_T16_Surf_83',
       '4435_Sta3_T20_Surf_87', '4439_Sta3_T24_Surf_91',
       '4443_Sta3_T28_Surf_95', '4447_Sta3_T32_Surf_99',
       '4451_Sta3_T36_Surf_103', '4455_Sta3_T40_Surf_107',
       '4459_Sta3_T44_Surf_111', '4463_Sta3_T48_Surf_115', 'Group'],
      dtype='object')

In [7]:

    

# sum over OGs for each timepoint to find out which timepoint is low: T0!
df_pro.ix[:,0:13].sum(axis=0)


    

    
Out[7]:





4415_Sta3_T0_Surf_20        53
4419_Sta3_T4_Surf_71      1992
4423_Sta3_T8_Surf_75      2005
4427_Sta3_T12_Surf_79     2063
4431_Sta3_T16_Surf_83     2028
4435_Sta3_T20_Surf_87     2077
4439_Sta3_T24_Surf_91     2025
4443_Sta3_T28_Surf_95     1977
4447_Sta3_T32_Surf_99     1884
4451_Sta3_T36_Surf_103    2119
4455_Sta3_T40_Surf_107    2026
4459_Sta3_T44_Surf_111    2036
4463_Sta3_T48_Surf_115    2043
dtype: int64

In [8]:

    

df_pro.drop('4415_Sta3_T0_Surf_20', axis=1, inplace=True)


    

In [9]:

    

# sum over timepoints for each OG to get total number of timepoints observed in
df_pro['Sum'] = df_pro.ix[:,0:13].sum(axis=1)


    

In [10]:

    

df_pro.Sum.value_counts()


    

    
Out[10]:





0     7136
12    1832
11      83
1       61
10      35
2       35
8       27
5       23
7       22
4       22
9       19
6       18
3       15
Name: Sum, dtype: int64

In [11]:

    

df_pro_nonzero = df_pro[df_pro.Sum > 0]


    

In [12]:

    

sns.boxplot(x=df_pro_nonzero.Group, y=df_pro_nonzero.Sum)


    

    
Out[12]:





<matplotlib.axes._subplots.AxesSubplot at 0x10dabcef0>






    

In [15]:

    

sns.kdeplot(data=df_pro.Sum[df_pro.Group == 'RedSea'], label='Red Sea')
sns.kdeplot(data=df_pro.Sum[df_pro.Group == 'Other'], label='Other')
plt.axis([-1.99, 13.99, 0, 0.4])
plt.xlabel('Number of timepoints observed in (out of 12)')
plt.ylabel('Proportion of gene ortholog groups')


    

    
Out[15]:





<matplotlib.text.Text at 0x10eeb04e0>






    

In [ ]: