In [1]:

    

from pandas import DataFrame


    

In [2]:

    

STITCH_TO_DRUGBANK = DataFrame.from_csv("./temp/chemical.sources.v5.0_DrugBank.tsv", sep='\t',
                                        index_col=None, header=None)
STITCH_TO_DRUGBANK = STITCH_TO_DRUGBANK.rename(columns={0: 'stitch_stereo', 1: 'stitch_flat', 2: 'database',
                                                 3: 'mapped_id'})


    

In [3]:

    

stereo_to_drugbank = list(set(zip(STITCH_TO_DRUGBANK.stitch_stereo,
             STITCH_TO_DRUGBANK.mapped_id)))
flat_to_drugbank = list(set(zip(STITCH_TO_DRUGBANK.stitch_flat,
             STITCH_TO_DRUGBANK.mapped_id)))


    

In [4]:

    

len(flat_to_drugbank)


    

    
Out[4]:





6923

In [5]:

    

unique_stereo_to_drugbank = list()

for element in stereo_to_drugbank:
    if element[0] not in [i[0] for i in unique_stereo_to_drugbank]:
        unique_stereo_to_drugbank.append(element)


    

In [7]:

    

prepare_for_insert = DataFrame(unique_stereo_to_drugbank, columns=['stitch_id', 'drugbank_id'])
prepare_for_insert = prepare_for_insert.set_index('stitch_id')


    

In [9]:

    

from sqlalchemy import create_engine

engine = create_engine('postgresql://postgres:postgres@localhost:32768/efktr')
prepare_for_insert.to_sql('stitch_to_drugbank', engine, if_exists='append')