#!/usr/bin/env python
# coding: utf-8

# In[52]:


import os
import pandas


# In[53]:


# downloaded from http://django.nubic.northwestern.edu/fundo/media/data/do_lite.txt
dolite = pandas.read_csv('do_lite.txt', sep='\t', names=['disease_name', 'gene_entrez', 'gene_symbol'])
dolite[:5]


# In[54]:


diseases = sorted(set(dolite['disease_name']))
len(diseases)


# In[ ]:





# In[55]:


with open('dolite_terms.txt', 'w') as write_file:
    write_file.write('\n'.join(diseases))


# In[56]:


path = os.path.join('..', 'data', 'term-names.tsv')
donames = pandas.read_csv(path, sep='\t')
donames[:5]


# In[57]:


donames['name_lower'] = [x.lower() for x in donames.name]
doname_map = donames[['doid', 'name_lower']].drop_duplicates()
doname_map[:5]


# In[58]:


dolite_df = pandas.DataFrame(data = diseases, columns = ['dolite_name'])
dolite_df['name_lower'] = [x.lower() for x in dolite_df.dolite_name]
dolite_df[:5]


# In[59]:


mapping_df = dolite_df.merge(doname_map, how='left')
mapping_df = mapping_df[['doid', 'dolite_name']].drop_duplicates()
mapping_df.to_csv('dolite_to_doid.tsv', sep='\t', index=False)
mapping_df[:5]


# In[60]:


# number of matches
sum(isinstance(x, str) for x in mapping_df.doid)


# In[ ]:





# In[ ]:





# In[ ]: