jhylin / data_in_life_blog Goto Github PK

# Query chembl_downloader to show SQL required 
# to extract ChEMBL data for a specific protein target
# e.g. target_chembl_id for acetylcholinesterase (AChE): CHEMBL220
queries.markdown(queries.get_target_sql(target_id="CHEMBL220", target_type="SINGLE PROTEIN"))

# Generate SQL for a query on the given target_id
# for acetylcholinesterase (AChE): CHEMBL220
sql = queries.get_target_sql(target_id="CHEMBL220", target_type="SINGLE PROTEIN")

# Pretty-print the SQL in Jupyter
queries.markdown(sql)

df = chembl_downloader.query(sql)

...

# Save df as .csv file (uncomment last line to run)
#df.to_csv("chembl_d_ache", sep=",", index=False)

...

# Load dataset from saved .csv file
df_ache = pd.read_csv("chembl_d_ache")
print(df_ache.shape)
df_ache.head()

from pathlib import Path

# Pick any directory, but make sure it's relative to your home directory
directory = Path.home().joinpath(".data", "blog")
# Create the directory if it doesn't exist
directory.mkdir(exist_okay=True, parents=True)

# Create a file path that corresponds to the version, since this could change
path = directory.joinpath(f"chembl_d_ache_{latest_version}.tsv")

if path.is_file():
    # If the file already exists, load it
    df_ache = pd.read_csv(path, sep=',')
else:
    # If the file doesn't already exist, make the query then cache it
    df_ache = chembl_downloader.query(sql)
    df_ache.to_csv(path, sep=",", index=False)