Integration of species information into the mapping data from NB

Goal: get proportion of reads from each KO by species



In [1]:

    
import pandas as pd
import matplotlib as mpl

import matplotlib.pyplot as plt
import cPickle as cpk


%matplotlib inline



In [2]:

    
#Load PhytoKEGG Annotations
AllPhytoKO_ann=pd.read_table('AllPhytoKegg_annotated.tab', header=False, delimiter='\t')



In [63]:

    
InsituCounts=pd.read_table('Data/AllInsitu_NoZero.tab', index_col='gID')



In [ ]:



In [64]:

    
#normalize to the library size
InsituTPM=InsituCounts.copy()
InsituTPM[['S1', 'S2', 'S3', 'S4', 'S5']]=(InsituCounts[['S1', 'S2', 'S3', 'S4', 'S5']]/InsituCounts[['S1', 'S2', 'S3', 'S4', 'S5']].sum())*10**6

#Add annotation information
InsituCounts=InsituCounts.join(AllPhytoKO_ann)
InsituTPM=InsituTPM.join(AllPhytoKO_ann)
InsituCounts=InsituCounts.dropna()
InsituTPM=InsituTPM.dropna()



In [6]:

    
#load in the species/group information
Group_Species=pd.read_table('GrpSpecies',delimiter=' ').T.drop(['MMETSP', 
                                                  'MMETSP.1']).T.drop_duplicates().set_index('SName')



In [21]:

    
InsituTPMGrped=InsituTPM.groupby(['kID','sgID']).sum().reset_index().set_index('sgID')



In [12]:

    
Group_Species=Group_Species.reset_index()



In [40]:

    
Dia=Group_Species[Group_Species['Grp']=='Bacillariophyta']
Din=Group_Species[Group_Species['Grp']=='Dinophyta']
Oth=Group_Species[((Group_Species['Grp']!='Dinophyta')&
                  (Group_Species['Grp']!='Bacillariophyta'))]



In [50]:

    
Insitu_TPM_DIA=InsituTPMGrped.loc[Dia['SName']].groupby('kID').sum()
Insitu_TPM_DIN=InsituTPMGrped.loc[Din['SName']].groupby('kID').sum()
Insitu_TPM_Oth=InsituTPMGrped.loc[Oth['SName'].dropna()].groupby('kID').sum()



In [59]:









    Out[59]:





S1    0.768644
S2    0.718057
S3    0.656064
S4    0.770223
S5    0.645120
dtype: float64



In [ ]: